Objectives:

Though there are numerous studies examining local control (LC) and overall survival (OS) following stereotactic ablative radiation therapy (SABR) for extracranial metastases in adult malignancies, reports on clinical outcomes for pediatric and adolescent/young adult (AYA) populations/histologies treated with SABR for either local recurrences or metastatic disease are limited. We aimed to address this gap through a multi-institutional analysis.

Methods:

The Radiosurgery Society (RSS) Patient Registry was queried for pediatric/AYA patients with extracranial metastases managed with SABR. AYA patients with adult histologies/primaries (ie. prostate adenocarcinoma, breast cancers, and cervical/endometrial cancers) were excluded. Kaplan-Meier analyses were employed to characterize LC and OS following SABR. Potential prognostic factors were assess with log-rank tests for initial univariate analysis (UVA). For multivariate analyses (MVA), a Cox proportional hazards multivariate model was utilized to further assess independent variables that were initially identified with a forward selection model.

Results:

We identified a total of 19 pediatric/AYA patients with 39 lesions treated with SABR available for analysis. Four lesions (10.3%) were treated for primary tumor recurrence, with the remainder treated for metastatic disease. The most common histologies of treated lesions were sarcomas (18 lesions; 46.1%, including chondrosarcoma, alveolar soft part sarcoma, and osteosarcoma) and central nervous system (CNS) malignancies spread to the spine (5 lesions; 12.8%, including anaplastic ependymoma and astrocytoma). The median patient age was 34 years (range: 16-39) with a median Karnofsky Performance Score (KPS) of 90% (range: 60-100). Sites of treated lesions included the lung (11; 28.2%), non-spinal bone (9; 23.1%), and spine (6; 15.4%) with a median gross tumor volume (GTV) of 12.4 cc (range: 0.7-233.5). The median prescription dose was 35 Gy (range: 12-60) delivered in a median of 5 fractions (range: 1-5) with a median biologically effective dose (BED10) of 61.5 Gy (range: 26.4-180). One-year LC and OS following SABR were 77.7% (95% CI: 58.5-88.7) and 72.7% (95% CI: 46.3-87.6), respectively. On UVA, BED10 ≥ 60 Gy was associated with significantly improved 1-year LC (94.4% vs. 47.6%; p < 0.0001) as were sarcoma primaries (2-year LC of 92.3% vs. 42.2%; p = 0.0002). CNS primaries had significantly poorer 1-year LC (87.5% vs. 20%; p < 0.0001) as well as spinal metastases (33.3% vs. 87.0%; p < 0.0001). Of note, 5/6 spinal metastases were from CNS primaries, with all having local failures. On MVA, including BED10, spinal metastases, and sarcoma histologies, BED10 < 60 Gy was associated with inferior LC (hazard ratio (HR) = 5.51; p = 0.01) with sarcoma primaries associated with superior LC (HR = 0.04; p = 0.008). On UVA, no prognostic factors were identified with respect to OS. No treatment related toxicities were reported.

Conclusion(s):

SABR with BED10 ≥ 60 Gy resulted in durable LC for pediatric and AYA patients with metastatic disease, though with poor outcomes noted in metastatic CNS malignancies regardless of treatment planning approach. Similar to recent findings in adult patients with metastatic disease from sarcoma primaries, SABR also has promising LC for sarcoma primaries and merits further investigation.