Session Item

UK SABR Consortium guidelines [1] recommend evaluating SABR plan quality using modified gradient index (MGI, or R50%). Its constraints are derived from plans analysed during the UK’s national SABR Commissioning through Evaluation (CtE) programme, and those observed by Yaparpalvi et al [2]. The constraints are based on PTV volume, but are defined in coarse increments. This results in significantly different target R50% for PTV volumes of 19.9cc and 20.1cc.

A planning study was performed in Pinnacle³ v16.4.3 using the CIRS lung phantom, investigating the effect of PTV volume, PTV oblate/prolateness, and lung density on the resulting R50%. VMAT plans were produced to local departmental practice, with Personalised Planning producing dual 210° arcs calculated at 2mm resolution. GTVs were defined using density overrides and were expanded by 5mm to form PTVs which ranged from 5 to 50cc, GTV oblateness was defined by major/minor axis ratio (a/b) from 0.5 to 2 (see figure), and lung density ranged from 0.15 to 0.35gcm^-3.

In line with [1] and [2], a relationship was found between PTV volume and R50%, but was acquired with finer increments than the UK national guidance.

Prolate targets (with respect to the cranio-caudal axis) had a significantly higher predicted R50% compared to spherical or oblate geometries, for the same volume. For an 18cc PTV, prolate (a/b =2) geometries had an average 10% larger R50% compared to oblate (a/b = 0.5). This relationship is more pronounced for larger PTV volumes as the PTV oblateness more closely matches that of the GTV.

Lung density has a significant impact on R50% due to wider scatter in low density media; a lung density of 0.2g/cm³ resulted in an 11% increase in R50% compared to the same 18cc PTV in a 0.3g/cm³ lung. Very low densities resulted in much larger R50%.

Our simple study demonstrates R50% is required at a finer volume resolution compared to the UK SABR Consortium guidance, but also that additional considerations are needed when assessing plan quality to account for PTV geometry and lung density.

This helps planners assess whether SABR plans are within expected ranges for given target volume, geometry and surrounding tissue, and reduce the likelihood of over-optimising the plan to meet the R50% target, or under-optimising a plan which could be improved.

References

[1] UK SABR Consortium, Stereotactic ablative body radiotherapy (SABR): a resource, v6.1, January 2019, available https://www.sabr.org.uk/wp-content/uploads/2019/04/SABRconsortium-guidelines-2019-v6.1.0.pdf

 [2] Yaparpalvi R, Garg MK, Shen J, Bodner WR, Mynampati DK, Gafar A, et al. Evaluating which plan quality metrics are appropriate for use in lung SBRT. Br J Radiol 2018; 91: 20170393