2051 - Preoperative Circulating Cell-Free DNA as a Prognostic Biomarker in Surgically Treated Spinal Metastatic Disease
Presenter(s)
K. Dibs1, R. Raval1, S. Zhu2, J. C. Grecula1, E. M. Thomas3, R. Singh1, F. Fekrmandi4, S. Beyer1, D. Handley5, J. B. Elder6, D. Schnieder7, B. Park8, D. M. Blakaj1, A. Chakravarti1, V. Chakravarthy8, and J. D. Palmer1; 1Department of Radiation Oncology, James Cancer Hospital/Wexner Medical Center, The Ohio State University, Columbus, OH, 2University of Florida, Gainesville, FL, 3The Renaissance Institute of Precision Oncology & Radiosurgery, Winter Park, FL, 4Department of Radiation Medicine, Roswell Park Comprehensive Cancer Center, Buffalo, NY, 5Center for Biostatistics, Department of Biomedical Informatics, The Ohio State University Wexner Medical Center, Columbus, OH, 6Ohio State University, Columbus, OH, 7The Ohio State University, Columbus, OH, 8Department of neurosurgery, The James Cancer Center, Ohio State University Wexner Medical Center, Columbus, OH
Purpose/Objective(s): Prognostic assessment for patients undergoing surgery for spinal metastatic disease is largely guided by clinical frameworks incorporating neurologic status, tumor biology, mechanical stability, and systemic disease burden. However, biologic markers reflecting real-time tumor activity are not routinely integrated into surgical decision-making. Circulating cell-free DNA (cfDNA) may provide complementary prognostic information. We evaluated the association between prospectively collected preoperative cfDNA levels and overall survival (OS) in patients undergoing surgery for spinal metastases.
Materials/Methods: In a prospective observational cohort, cfDNA samples were collected at predefined perioperative time points. cfDNA was quantified using next-generation sequencing (NGS) [ng/mL]. For this analysis, we included 40 patients who underwent spine surgery for metastatic disease and had available preoperative cfDNA measurements. cfDNA levels were dichotomized at 14 ng/mL. OS was calculated from the date of surgery using Kaplan–Meier methods and compared with log-rank testing. Univariable and multivariable Cox proportional hazards models evaluated associations between clinical variables and OS.
Results:
At last follow-up, 13 patients were alive. Median age was 70 years (range 46–85), 58% were male, and 67% had ECOG 0–1. Fifty-five percent had bone-only disease. On Kaplan–Meier analysis, systemic disease extent significantly impacted OS, with inferior survival observed in patients with bone plus visceral metastases (p=0.005). Age (p=0.12), ECOG (p=0.63), sex (p=0.88), and concurrent systemic therapy (p=0.55) were not significantly associated with OS.
Patients with preoperative cfDNA =14 ng/mL demonstrated significantly worse OS compared with <14 ng/mL (p=0.0097). On univariable Cox analysis, bone plus visceral disease (HR 2.88, 95% CI 1.33–6.24, p=0.007) and elevated cfDNA (HR 2.76, 95% CI 1.24–6.11, p=0.013) were associated with increased mortality. On multivariable analysis adjusting for age, sex, ECOG, and systemic therapy, both bone plus visceral disease (HR 2.89, 95% CI 1.17–7.15, p=0.022) and elevated preoperative cfDNA (HR 2.62, 95% CI 1.10–6.23, p=0.029) remained independently associated with worse OS. Although longitudinal cfDNA levels demonstrated a significant temporal trend (p=0.016), change over time was not correlated with OS (R=-0.07, p=0.68).
Conclusion:
In this prospective cohort, preoperative cfDNA provided independent prognostic information beyond established clinical factors in surgically treated spinal metastatic disease. Integration of biologic biomarkers with systemic disease assessment may enhance risk stratification and inform multidisciplinary surgical decision-making. Given the limited sample size (40 patients with 27 events), these findings require validation in larger cohorts to mitigate potential model overfitting. Prospective validation in larger cohorts is warranted.