Csi Planning

Kearla Bentz CSI Planning 1
Contours and Fields

For the CSI plan I explored the internet to find a planning technique to use. I found a study
by Prabhu et al1 that gave instructions for volumetric modulated arc therapy (VMAT) planning
that used a “gradient-optimized” method. Following the guidance of the study, I contoured each
humerus and added a 10 mm expansion to generate a planning organ at risk volume (PRV) and
minimize dose delivery through the arms. I also divided the planning target volume (PTV) spine
into 3 regions: upper, middle, and lower thecal sac to help with coverage in the optimizer.
Prabhu et al1 suggested field borders that overlap by 4 cm at the depth of the PTV spine and
the regions of overlap be subdivided into 4 equal structures (A1-A4 and B1-B4 with 2 spinal
fields). To do this I used the “Boolean Operators” feature in contouring and created a zPTV_all
structure that combined the PTV cranial with the PTV spine. I then used the “Arc Geometry
Tool” in the Eclipse version 15.6 treatment planning system (TPS) to place the isocenters. I
selected the zPTV_all as my target and 3 isocenters for the setup and then I fit the collimator to
the target.
After the isocenters were created by Eclipse, I added additional arcs so that each
isocenter had 2 coplanar arcs with collimator angles of 10° and 350°. I matched the technique of
Prabhu et al1 and used 6 MV beams for each arc and had the dose rate at 600. The fields did not
overlap by 4 cm exactly, but I subdivided the overlapping regions with 4 equal structures (A1-A4
and B1-B4).
Optimization
For dose calculation of this CSI VMAT plan AcurosXB version 15.6 was used. Before
optimization the calculation options in the photon optimizer were customized. For this plan auto-
feathering was turned on. This feature superimposes a linear low-dose gradient between
isocenter beams.2
In the optimizer I ran multiple structures with upper and lower objectives (pictured
below). I also put a “no entry” option on the bilateral humerus PRV structures to avoid shooting
beams through the arms. Additionally, the structures A2, A3, B2, B3 were used to help fill in
dose and eliminate hot spots in the areas of overlap. This process did not completely follow the
methods of Prabhu et al1 because I chose to optimize the 3 isocenters simultaneously, rather than
optimizing the isocenters separately and using the middle isocenter plan as a base dose for the
upper and lower isocenters.
Normalization
After the plan was optimized and calculated, I evaluated the plan to assess PTV coverage
and dose constraints before normalization. I re-optimized the plan a few times to meet dose
constraints and improve PTV coverage. I normalized this plan to 99.4% manually because
without normalization the PTV cranial was slightly below the goal of V34.2 Gy ≥ 95% when
uploaded into ProKnow.
Results
The final results of my CSI VMAT plan are shown below. This plan was evaluated based
on the ProKnow performance bin scoring constraints. I was unable to meet the left and right
kidney mean dose constraints of ≤ 3 Gy. I tried to reduce the kidney mean dose to ≤ 3 Gy, but I
noticed that the PTV spine coverage was being affected so I quit pushing so hard on it in the
optimizer. I made sure that I still met the kidney mean constraint in the margin region for
ProKnow scoring. Below are links to a Kaltura capture that shows the axial and sagittal isodose
coverage. Additionally, the ProKnow constraints and axial images of the plan showing the
isodose coverage are shown below. The axial images start superiorly and progress inferiorly in 3
cm increments.
https://cdnapisec.kaltura.com/index.php/extwidget/preview/partner_id/2370711/uiconf_id/42909
941/entry_id/1_g9g99qbw/embed/dynamic
https://cdnapisec.kaltura.com/index.php/extwidget/preview/partner_id/2370711/uiconf_id/42909
941/entry_id/1_3jii5zhn/embed/dynamic
The maximum dose for this plan was 109.2% (39.31 Gy). It was located in the overlap
region of the upper isocenter and the middle isocenter fields in the posterior spinous process
area. At my current clinical facility this maximum dose percentage and location would be
acceptable.
This plan did have some areas of PTV that were not receiving prescription dose. This
occurred near the kidneys from trying to meet the kidney mean dose constraint of ≤ 3 Gy. The
minimum dose for the PTV spine was 35.6% (12.82 Gy). At my clinical site the kidney mean
constraint would be less of a priority than the PTV coverage, allowing a conformal dose to the
PTV with a reasonable dose to the kidneys. This would increase the minimum dose to the PTV
and avoid cold spots within the PTV.
Discussion
I enjoyed planning this VMAT CSI treatment. I had never used the “Arc Geometry Tool”
before for a multiple isocenter plan. I thought that having this tool was helpful, and it saved a
bunch of time by not having to optimize 3 different isocenters separately. I would like to know if
there was a way to meet the kidney mean constraint of ≤ 3 Gy without compromising PTV
coverage. If I had to replan this treatment using the same technique that I used, I would let the
kidney mean dose be around 5 Gy to see if that helped with PTV coverage.
References
1. Prabhu RS, Dhakul R, Piantino M, et al. Volumetric modulated arc therapy (VMAT)
craniospinal irradiation (CSI) for children and adults: a practical guide for
implementation. Practical Radiat Oncol. 2022;12(2):e101-e109.
https://doi.org/10.1016/j.prro.2021.11.005
2. Matsumoto K, Monzen H, Kubo K, Otsuka M, Nambu H, Nishimura Y. Volumetric
modulated arc therapy planning for craniospinal irradiation with a new o-ring linac.
Cureus. 2023;15(3):e36493. https://doi.org/10.7759/cureus.36493

Csi Planning

Uploaded by

Copyright:

Available Formats

Csi Planning

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Csi Planning

Uploaded by

Copyright:

Available Formats

Kearla Bentz CSI Planning 1

Contours and Fields

You might also like