Advanced

Accuracy of MLC-tracking for inversely optimized arc therapy treatments of varying complexity for two MLCs

Larsson, Tobias (2010)
Medical Physics Programme
Abstract (Swedish)
En mycket vanlig behandlingsteknik vid cancer är strålbehandling. Tekniken har utvecklats snabbt de senaste åren och idag uppnås millimeterprecision för många behandlingsområden. Därmed har behovet av att kunna kompensera för rörelser under behandlingen ökat. I detta arbete undersöktes möjligheten att låta strålgången följa tumören med en teknik som kallas MLC-tracking.

Strålbehandling började användas bara något år efter Wilhelm Röntgens upptäckt av röntgenstrålning och har därefter utvecklats till att vara en mycket avancerad behandlingsform. Framförallt de senaste 20 åren har inneburit en enorm utveckling inom området så att röntgenstrålning med hög energi kan riktas mot en tumör med millimeterprecision, samtidigt som bestrålning av... (More)
En mycket vanlig behandlingsteknik vid cancer är strålbehandling. Tekniken har utvecklats snabbt de senaste åren och idag uppnås millimeterprecision för många behandlingsområden. Därmed har behovet av att kunna kompensera för rörelser under behandlingen ökat. I detta arbete undersöktes möjligheten att låta strålgången följa tumören med en teknik som kallas MLC-tracking.

Strålbehandling började användas bara något år efter Wilhelm Röntgens upptäckt av röntgenstrålning och har därefter utvecklats till att vara en mycket avancerad behandlingsform. Framförallt de senaste 20 åren har inneburit en enorm utveckling inom området så att röntgenstrålning med hög energi kan riktas mot en tumör med millimeterprecision, samtidigt som bestrålning av frisk vävnad kan undvikas i större och större utsträckning.

Att tumören rör sig under behandlingen är däremot ett problem som fortfarande innebär stora utmaningar. Det är framförallt vid behandling i området kring lungorna som andningen orsakar omfattande tumörrörelser. Lösningen är att ett större område bestrålas, så att strålningen alltid träffas tumören. Nackdelen med detta är att en större del av den friska vävnaden bestrålas och därmed ökar risken för biverkningar.

Ett alternativ till att utöka det bestrålade området är att låta strålfältet följa tumören under behandlingen. Detta görs genom att den anordning som formar strålfältet, flerbladskollimatorn (MLCn), i realtid justeras efterhand som tumören rör sig. Flerbladskollimatorn består av metallblad som tillsammans formar strålfältet. Tekniken att låta strålningen följa tumören kallas MLC-tracking och i detta arbete undersöktes hur väl MLC-tracking fungerar för en behandlingsform som kallas RapidArc. RapidArc kännetecknas av att bestrålningen sker samtidigt som behandlingsapparaten roterar runt patienten och flerbladskollimatorn formar strålfältet.

Totalt undersöktes 44 olika behandlingsplaner av varierande komplexitet för två olika typer av flerbladskollimatorer. Resultaten visar att MLC-tracking har potentialen att mycket effektivt kompensera för rörelser, men att prestandan försämras för mer avancerade behandlingsplaner. Resultatet visar också att avstånden mellan bladen i flerbladskollimatorn är avgörande för hur väl tumörrörelser kan kompenseras. (Less)
Abstract
"Purpose: To investigate the geometric accuracy of MLC-tracking using a circular field, and to investigate the dosimetric accuracy for inversely optimized arc delivery to a moving target. This is done
for two different MLCs and for varying plan complexity.

Materials and methods: For investigating the geometric accuracy, a marker was placed on a motion platform while the MLC set to shape a circular field. During tracking of the target, EPID imaging was performed and the geometric accuracy, i.e. the difference between the marker and the center of the MLC shape, was calculated for different assumed system latencies. To investigate the dosimetric accuracy, plans were made in Eclipse™ treatment planning system using the RapidArc® treatment... (More)
"Purpose: To investigate the geometric accuracy of MLC-tracking using a circular field, and to investigate the dosimetric accuracy for inversely optimized arc delivery to a moving target. This is done
for two different MLCs and for varying plan complexity.

Materials and methods: For investigating the geometric accuracy, a marker was placed on a motion platform while the MLC set to shape a circular field. During tracking of the target, EPID imaging was performed and the geometric accuracy, i.e. the difference between the marker and the center of the MLC shape, was calculated for different assumed system latencies. To investigate the dosimetric accuracy, plans were made in Eclipse™ treatment planning system using the RapidArc® treatment technique for two lung cancer patients with a prescribed dose of 2 Gy per fraction. A gantry rotation span of 225° to 135° was used. Four sets of increasingly stringent planning dose objectives (PO) were applied in planning for delivery on a Novalis TX™ linear accelerator (equipped with a High definition MLC (HDMLC)) and on a Varian 2300ix linear accelerator (with the Millennium 120 MLC), for two collimator angles (CA); 45° and 88°. For each patient, one plan was created with CA45 and with an optimization constraint limiting the distance to adjacent leaves. Plans were also created in a clinical version of Eclipse for the PO#1 and PO#2 sets. The number of MU ranged from 334 to 751 for all plans. The plans were delivered to a Delta4® dosimetric device placed on a motion platform moving sinusoidally in the SI direction, with 2.0 cm peak to peak motion and 6 s cycle time. Position monitoring was done using the ExacTrac® optical system. Measurements were made with and without MLC-tracking, and with and without motion. The measurements with a stationary target were used as reference in gamma evaluation, with gamma criteria of 2% and 2 mm (using a dose region of 5-500%). The calculated dose distributions were also used as reference, with gamma criteria of 3% and 3 mm.

Results: For the most suitable assumed latency, the geometric accuracy expressed as RMS value was 0.316 mm and 0.346 mm for the HDMLC and the Millenium MLC respectively. For all RapidArc plans measured, the MLC-tracking method considerably increased the gamma index pass rate for delivery to a moving target compared to delivery with no motion compensation (using the measured dose on a static phantom a reference). The pass rate was also improved for CA 88° compared to 45°.
The gamma index pass rate decreased with #MU for both MLCs with CA 45°. Dose profile analyses showed that overdosage in the high dose region was the primary cause of gamma evaluation failures. The pass rate was significantly higher (P=0.0025) for measurements using the Millennium MLC compared to the HDMLC. A correlation was seen between reduced average adjacent leaf distance (weighted against the dose weight for the corresponding control point) and improved gamma index pass rate.

Conclusion: It is possible to use MLC-tracking during RapidArc® delivery to compensate for the target motion simulated in this study. The gamma index pass rate was increased using MLC-tracking, but the effect tended to decrease slightly for the more complex plans. Aligning the leaves with the target motion substantially increased the gamma index pass rate for both MLCs, regardless of the plan complexity. The distance to adjacent MLC leaves seems to be an important factor in predicting MLC-tracking performance." (Less)
Please use this url to cite or link to this publication:
author
Larsson, Tobias
supervisor
organization
year
type
H2 - Master's Degree (Two Years)
subject
keywords
Strålterapi
language
English
id
2157118
date added to LUP
2011-09-14 12:01:39
date last changed
2013-09-05 10:23:34
@misc{2157118,
  abstract     = {"Purpose: To investigate the geometric accuracy of MLC-tracking using a circular field, and to investigate the dosimetric accuracy for inversely optimized arc delivery to a moving target. This is done
for two different MLCs and for varying plan complexity.

Materials and methods: For investigating the geometric accuracy, a marker was placed on a motion platform while the MLC set to shape a circular field. During tracking of the target, EPID imaging was performed and the geometric accuracy, i.e. the difference between the marker and the center of the MLC shape, was calculated for different assumed system latencies. To investigate the dosimetric accuracy, plans were made in Eclipse™ treatment planning system using the RapidArc® treatment technique for two lung cancer patients with a prescribed dose of 2 Gy per fraction. A gantry rotation span of 225° to 135° was used. Four sets of increasingly stringent planning dose objectives (PO) were applied in planning for delivery on a Novalis TX™ linear accelerator (equipped with a High definition MLC (HDMLC)) and on a Varian 2300ix linear accelerator (with the Millennium 120 MLC), for two collimator angles (CA); 45° and 88°. For each patient, one plan was created with CA45 and with an optimization constraint limiting the distance to adjacent leaves. Plans were also created in a clinical version of Eclipse for the PO#1 and PO#2 sets. The number of MU ranged from 334 to 751 for all plans. The plans were delivered to a Delta4® dosimetric device placed on a motion platform moving sinusoidally in the SI direction, with 2.0 cm peak to peak motion and 6 s cycle time. Position monitoring was done using the ExacTrac® optical system. Measurements were made with and without MLC-tracking, and with and without motion. The measurements with a stationary target were used as reference in gamma evaluation, with gamma criteria of 2% and 2 mm (using a dose region of 5-500%). The calculated dose distributions were also used as reference, with gamma criteria of 3% and 3 mm.

Results: For the most suitable assumed latency, the geometric accuracy expressed as RMS value was 0.316 mm and 0.346 mm for the HDMLC and the Millenium MLC respectively. For all RapidArc plans measured, the MLC-tracking method considerably increased the gamma index pass rate for delivery to a moving target compared to delivery with no motion compensation (using the measured dose on a static phantom a reference). The pass rate was also improved for CA 88° compared to 45°.
The gamma index pass rate decreased with #MU for both MLCs with CA 45°. Dose profile analyses showed that overdosage in the high dose region was the primary cause of gamma evaluation failures. The pass rate was significantly higher (P=0.0025) for measurements using the Millennium MLC compared to the HDMLC. A correlation was seen between reduced average adjacent leaf distance (weighted against the dose weight for the corresponding control point) and improved gamma index pass rate.

Conclusion: It is possible to use MLC-tracking during RapidArc® delivery to compensate for the target motion simulated in this study. The gamma index pass rate was increased using MLC-tracking, but the effect tended to decrease slightly for the more complex plans. Aligning the leaves with the target motion substantially increased the gamma index pass rate for both MLCs, regardless of the plan complexity. The distance to adjacent MLC leaves seems to be an important factor in predicting MLC-tracking performance."},
  author       = {Larsson, Tobias},
  keyword      = {Strålterapi},
  language     = {eng},
  note         = {Student Paper},
  title        = {Accuracy of MLC-tracking for inversely optimized arc therapy treatments of varying complexity for two MLCs},
  year         = {2010},
}