254 / 1096
S241
ESTRO 36
_______________________________________________________________________________________________
Purpose or Objective
Published data demonstrate an overall survival benefit
associated with including the internal mammary chain
(IMC) in the radiotherapy target volume (TV) for women
with node positive breast cancer. Implementation of IMC-
RT will be facilitated by development of resource efficient
techniques. However, even relatively simple techniques
rely on time consuming clinician outlining of lymph nodes
to achieve adequate dose to the TV (not well covered by
standard fields based on bony landmarks). In order to
reduce the resource burden of nodal contouring, an
anatomical point based algorithm for guiding field
placement was developed and tested for its ability to
ensure TV coverage.
Material and Methods
We identified six points, representative of regional lymph
node level borders according to ESTRO consensus
guidelines, and tested these for their ability to inform
field placement adequately covering the TV (Table 1).
Written instructions were developed and a cohort of 20
cases identified as a validation group. ‘Gold standard’
nodal volumes (Levels 1-4 and IMC) were delineated
according to ESTRO consensus guidelines by four clinical
oncologists with experience in locoregional breast
radiotherapy. Six independent testers (three clinicians
and three radiographers blinded to the nodal volumes)
were invited to place points and consequently fields on
the cases. In four cases a clinician placed both the points
and fields, in eight cases a clinician placed the points and
radiographer applied the fields and in eight cases points
and fields were placed by a radiographer. Cases were
planned using forward planned techniques. The dose
objective to the nodal PTV was V
32Gy
≥90%.
Results
Fourteen of 20 cases met the dose objectives when testers
followed the written algorithm alone (Figure 1). Of the
remaining six cases, four failed due to the subclavian vein
being mistaken for the subclavian artery. Two failed due
to point 3 being placed at the inferior part of the
pectoralis minor muscle resulting in insufficient coverage
of level 3. When the points were re-placed correctly nodal
TVs were well covered.
Conclusion
The results suggest that, in the majority of cases, by
following the algorithm clinicians and radiographers can
appropriately place fields which result in acceptable nodal
TV coverage. To avoid the common mistakes encountered,
a new approach to training including verbal instructions is
being developed for testing. Our aim is to demonstrate
benefits in clinical workflow and expand validation to
other RT centres.
OC-0453 Stereotactic radiosurgery for multiple brain
metastases: Results of multi-centre benchmark studies
D.J. Eaton
1
, J. Lee
1
1
Radiotherapy Trials QA group RTTQA, Mount Vernon
Hospital, Northwood, United Kingdom
Purpose or Objective
Stereotactic radiosurgery (SRS) is strongly indicated for
treatment of multiple brain metastases. Various
treatment platforms are available, and comparisons have
been made between modalities, but mostly in single
centre studies. A pre-requisite for all providers selected
as SRS/SRT centres in England was to participate in a
quality assurance process, informed through collaboration
between the national trials QA group and a
multidisciplinary expert advisory group. All clinical
centres undertook planning benchmark cases, providing a
unique dataset of current practice across a large number
of providers and a wide range of equipment. This was used
to facilitate sharing of best practice and support centres
with less experience.
Material and Methods
Two brain metastases cases were provided, wit h images
and structures pre-drawn, involving three and seven
lesions respectively. Centres produced plans a ccording to
their local practice, and these were reviewed centrally
using metrics for target coverage, selectivity, gradient
fall-off and normal tissue sparing.
Results
38 plans were submitted, using 21 differe nt treatment
platforms, including Gamma Knife, Cyberknife, Varian
(Novalis / Truebeam STx / 2100), Elekta (Synergy / Versa
HD using Beam Modulator / Agility MLC) and Tomotherapy.
6 plans were subsequently revised following feedback, and
review of 4 plans led to a restriction of service in 3
centres. Prescription doses were typically 18-25Gy in 1
fraction (or 27/3fr), except for a lesion within the
brainstem, which was prescribed 12-20Gy in 1 fraction (or
18-30Gy/5fr). All centres prioritised coverage, with the
prescription isodose covering ≥95% of 208/209 lesions.
Selectivity was much more variable, especially for smaller
lesions, and in some cases this was combined with high
gradient index, resulting in large volumes of normal tissue
being irradiated. Both Tomotherapy submissions were
outliers in terms of either selectivity or gradient index,
but all other platforms were able to give plans with
relatively low gradient indices for larger lesion volumes,
although there was more variation among Varian and
Elekta plans, than for Gamma Knife and Cyberknife (first
figure). There were also larger differences for the smaller
volumes, with increasing variation both inter- and intra-
treatment-platform. Doses to normal brain and brainstem
were highest when PTV margins were applied, but
substantial improvements were possible by re-planning,
even without changing margin size (second figure).