Dose Escalation in Stereotactic Body Radiation Therapy for Pancreatic Cancer: A Meta-Analysis.

Abstract

To determine whether increasing biologically effective dose (BED) with stereotactic body radiation therapy (SBRT) is associated with improved local control (LC) or toxicities in patients with locally advanced pancreatic cancer. A PICOS/PRISMA/MOOSE selection protocol was used to identify 15 studies across 12 institutions in 5 countries where patients received definitive SBRT for nonmetastatic disease. Biologically equivalent doses were calculated with an α/β of 10 (ie, BED10) for LC and acute toxicity and 3 (ie, BED3) for late toxicity. Fixed and random effects models were used to characterize LC and grade 3/4 toxicities by BED. There were 508 patients included with a median follow-up time of 9.1 months. The median dose was 30 Gy, and the most common regimen was 30 Gy/5 fractions. There was no significant difference in LC rates at 1 year between the BED10<70 Gy versus ≥70 Gy groups, with an estimate of 0.60 (95% confidence interval [CI], 0.36-0.81) versus 0.83 (95% CI, 0.63-0.97), respectively. There was no significant difference in acute toxicity rates between the BED10<70 Gy versus ≥70 Gy groups, with an estimate of 0.02 (95% CI, 0.00-0.08) versus 0.05 (95% CI, 0.00-0.22), respectively. Given the dose distribution across studies, 3 intervals were used to characterize BED3. There were no significant differences in late toxicity among those receiving BED3<100, 100 to 200, or >200 Gy. SBRT for pancreatic cancer results in LC rates of 60% to 83% and clinically significant toxicity of <7%. Increasing BED10 beyond 70 Gy was not associated with increased rates of 1-year LC or acute toxicity. Increasing BED3 beyond 100 Gy was not associated with increased rates of late toxicity.