SBRT in pancreatic cancer: what is the therapeutic window?

Abstract

To analyse outcome and toxicity of stereotactic body radiotherapy (SBRT) in pancreatic cancer (PDAC). We systematically reviewed full reports on outcome and toxicity transforming prescription doses to equivalent doses of 2 Gy (EQD2) and biological equivalent doses (BED). Pearson product-moment correlation coefficient, regression analysis and Lyman-Kutcher-Burman modelling were used. Sixteen trials (572 patients) were identified. Local control correlated with dose. Additionally 4 upper gastrointestinal-SBRT trials (149 patients) were included for toxicity analysis. Acute toxicity was mild but late toxicity ⩾G2 was substantial and predominantly gastrointestinal. Late toxicity ⩾G2 and ⩾G3 correlated highly with EQD2/BED after linear (R(2)=0.85 and 0.77, respectively) and Lyman-Kutcher-Burman modelling. Linear regression lines indicated ⩾G2 and ⩾G3 toxicity frequencies of 5% at 65 Gy and 80 Gy EQD2-α/β=3, respectively. A comparison of toxicity with dose constraints for duodenum revealed partly inadequate dose constraints. RESULTS from multiple fraction regimens could be successfully interpreted to estimate toxicity according to EQD2/BED prescription doses, and dose constraints for the duodenum were derived, whereas local control appeared to be less dose-dependent. This analysis may be useful to plan clinical trials for SBRT and hypofractionated radiotherapy in pancreatic cancer.