Main Session
Sep
29
PQA 07 - Head and Neck Cancer, Lung Cancer/Thoracic Malignancies, and Nursing and Supportive Care
3507 - Clinical Outcomes of Zero-Margin, Single Isocenter Radiosurgery for Brain Metastases from Small Cell vs. Non-Small Cell Lung Cancer
Presenter(s)
Nicholas Johnson, BS - University of Alabama at Birmingham, Birmingham, AL
L. Moradi1, N. R. Johnson2, R. A. Popple1, J. Hoyle1, C. D. Willey1, J. A. Pogue2, H. Boggs1, and J. B. Fiveash1; 1University of Alabama at Birmingham Department of Radiation Oncology, Birmingham, AL, 2University of Alabama at Birmingham, Birmingham, AL
Purpose/Objective(s):
Brain metastases from small cell lung cancer (SCLC) were historically excluded from early radiosurgery (SRS) trials, thus efficacy data is limited. Although SCLC is generally considered more radiosensitive than non–small cell lung cancer (NSCLC), its potential infiltrative growth pattern has raised concern that a 0-mm planning target volume (PTV) margin may inadequately cover microscopic tumor extent. This retrospective study compares local control and toxicity of SCLC versus NSCLC brain metastases treated with single-isocenter, multitarget, LINAC-based radiosurgery using a 0-mm PTV margin.Materials/Methods:
A retrospective analysis was conducted of 265 lung cancer patients with a total of 1300 brain metastases treated with SRS to compare local control and toxicity of SCLC vs. NSCLC histology. All treatment plans used a single prescription dose, generally determined by the largest lesion. Treatments were delivered using a 0 mm PTV margin with single-isocenter, LINAC-based radiosurgery. Local failure was defined as a =25% increase in maximum tumor diameter (minimum absolute increase of 3 mm) or histopathologic confirmation of viable tumor at salvage surgery. Toxicity was defined as grade =3 central nervous system adverse events per CTCAE v5.0. Local control and toxicity were estimated using the Kaplan–Meier method. Associations were evaluated using Cox proportional hazards regression. Analyses were conducted on a per-tumor basis.Results:
96 brain metastases with SCLC and 1204 brain metastases with NSCLC received SRS in this analysis.1-year local control for SCLC vs. NSCLC was 81% vs. 96% (p <0.001). 1-year toxicity for SCLC vs. NSCLC was not statistically different (99% vs. 98%, p = 0.733). On multivariable analysis, SCLC histology (HR 3.741, p<0.001), tumor volume (HR 1.092, p<0.001 were found to be independent predictors of local control. In a separate analysis of SCLC vs. NSCLC treated with 18 Gy, SCLC had a significantly worse 1-year local control (85% vs. 97%, p = 0.004). Histology did not predict local control if 20 Gy was prescribed.Conclusion:
In this cohort, local control of brain metastases from SCLC was worse than NSCLC when treated with 0 mm PTV margin, single isocenter, and multi-target radiosurgery. Future studies should investigate clinical factors, such as the dose regimen and target volume definition, to determine their effects on local control.