Prognostic Value of Primary Tumor Volume Changes on kV-CBCT during Definitive Chemoradiotherapy for Stage III Non-Small Cell Lung Cancer.

INTRODUCTION: Kilovoltge cone beam computed tomography (kV-CBCT) allows for tumor localization and response assessment during definitive chemoradiotherapy for locally advanced NSCLC. We hypothesize that significant tumor volume loss occurs early during radiotherapy and that the extent of volume loss correlates with clinical outcomes.

METHODS: A total of 52 patients with locally advanced NSCLC treated with definitive chemoradiotherapy were reviewed. kV-CBCT images were used to contour primary gross tumor volumes at four time points during treatment. Patients were dichotomized according to absolute and relative volume changes at each time point. Statistical analyses were performed to evaluate correlations between volume changes and clinical outcomes.

RESULTS: The median gross tumor volumes were 77.1, 48.3, 42.5, and 29.9 cm3 for fractions 1, 11, 21, and final, respectively. Greater relative volume loss between fractions 1 and 21 correlated with improved distant control (hazard ratio [HR] = 0.35, 95% confidence interval [CI]: 0.13-0.94, p = 0.038) and overall survival (HR = 0.40, 95% CI: 0.16-0.98, p = 0.046). Greater relative volume loss between fractions 11 and 21 correlated with improved progression-free survival (HR = 0.39, 95% CI: 0.17-0.88, p = 0.02) and trended toward improved overall survival (HR = 0.43, 95% CI: 0.17-1.06, p = 0.07). On multivariate analysis, greater relative volume loss between fractions 11 and 21 correlated with improved progression-free survival (HR = 0.39, 95% CI: 0.16-0.97, p = 0.041) and overall survival (HR = 0.31, 95% CI: 0.11-0.88, p = 0.027).

CONCLUSIONS: Significant primary tumor volume loss occurs early during radiotherapy for locally advanced NSCLC. Greater relative tumor volume loss during treatment correlates with improved disease control and overall survival. Thus, kV-CBCT has potential to be used as a practical prognostic imaging marker.