Dosimetric analysis of VMAT for locally advanced pancreatic cancer and a novel duodenal-PTV overlap parameter as a predictor of feasibility of dose escalation


Year:

Session type:

Theme:

Paul Junni1,Ganesh Radhakrishna1
1NHS

Abstract

Background

This study evaluates the use of conventional and dose escalated volumetric-modulated arc therapy (VMAT) compared with 3D-conformal radiotherapy for locally advanced pancreas cancer, and investigates whether a novel parameter of the overlap volume of the PTV with the duodenum (duoOLV) or stomach (stoOLV) can be used to predict in whom dose escalation is feasible.

Method

21 consecutive patients who had undergone 3D-conformal radiotherapy 54Gy in 30 fractions for locally advanced pancreatic cancer were replanned with VMAT to doses of 54Gy in 30 fractions and 59.4Gy in 33 fractions. Dosimetry to target volumes and organs at risk (OAR) along with irradiated volumes were compared.

Results

Compared with 3D-conformal radiotherapy, VMAT54 provided a significant reduction in dose to stomach, duodenum, small bowel and kidneys. VMAT59 significantly increased mean GTV and PTV doses with no loss of coverage of the 95% isodose from the 54Gy plans.   The duodenal constraint of V55<1cm3 was exceeded in 7, 0 and 0 of 3D-conformal, VMAT54 and VMAT59.4 plans respectively. The stomach constraint of V50<16cm3 was exceeded in 7, 6, and 2 of 21 patients with 3D-conformal, VMAT54 and VMAT 59.4 plans respectively. There was no significant difference in irradiated volume comparing VMAT54 and VMAT59.4 plans. For VMAT59.4 mean PTV and GTV dose and coverage of GTV by 95% isodose showed a significant correlation with duoOLV (p<0.001); a duoOLV >21cm3 predicted a failure to achieve 95% prescription dose coverage (V95%Px) for the GTV of >90%. No correlation was found with stoOLV and target volume coverage for VMAT59.4.

Conclusion

Compared with 3D-conformal radiotherapy, VMAT without dose escalation provides significant OAR sparing. Dose escalation to 59.4Gy with VMAT whilst respecting duodenal and stomach constraints is feasible in a subset of patients. A duoOLV of >21cm3 predicted a failure to achieve dose escalation whilst maintaining duodenal doses within tolerance.