Carbon ion radiotherapy optimization techniques for pancreatic cancer: accounting for the effect of bowel gas variation.

[BACKGROUND] Dose variation due to changes in bowel air poses significant challenges for carbon radiotherapy in pancreatic cancer. This retrospective study evaluated a density-override optimization technique to mitigate dosimetric uncertainties caused by bowel air changes.

[MATERIALS AND METHODS] Planning CT and cone-beam CT data from 8 patients with locally advanced pancreatic cancer undergoing stereotactic ablative radiotherapy were analyzed. Treatment simulations used a dose of 55.2 GyE in 12 fractions with a four-field setup (anterior, lateral, posterior, posterior oblique). Four density-override patterns were compared: pattern 0 (no override), pattern 1 (replacing bowel gas with water), pattern 2 (replacing the entire bowel with mean bowel HU), and pattern 3 (replacing bowel gas with mean bowel HU). Dose evaluations included fraction-wise and accumulated dose analyses, focusing on target coverage, homogeneity index, and organs at risk doses.

[RESULTS] Pattern 2 achieved the largest clinical tumor volume coverage and the fewest fractions with > 5% coverage loss for the anterior beam, followed by pattern 3. However, pattern 2 demonstrated poorer homogeneity for the lateral beam compared to patterns 1 and 3 and a higher gastrointestinal (GI) dose for the anterior beam.

[CONCLUSION] This study evaluated the importance of density overrides to address bowel air variations. For patients where a more uniform dose is desirable or whose tumor is adjacent to the GI tract, a pattern 3 density-override should be considered.