One such paper, presented by Marcel van Herk of the Netherlands Cancer Institute, examined the difficulties posed by differential motion between the target and the organs-at-risk (OAR). Van Herk explained that while current image-guidance systems can provide efficient correction of tumour position, if the other organs don't move in the same way, accurate target positioning may result in over-irradiation of these sites. To deal with such differential motion, he proposed an image-guided radiotherapy system in which OAR motion is considered alongside tumour motion during treatment planning and patient set-up. When tested on 86 patients with peripheral lung tumours, in all cases the scheme could deliver a positioning correction that satisfied both target and OAR constraints.

Dealing with motion is particularly demanding when it comes to complex treatments such as intensity-modulated radiotherapy (IMRT). "Moving targets are an important problem in the use of RapidArc and other intensity-modulated techniques," said Jens Zimmerman, from the Finsen Center, Rigshospitalet in Denmark. In another of the papers highlighted by the ESTRO Physics Committee, Zimmerman presented a study evaluating the dose delivered to a moving target from a Varian RapidArc system. "We can deal with this motion by monitoring the target and making the beam follow its movements in real time," he explained, adding that the beam tracking is achieved using a dynamic multileaf collimator (DMLC) controlled by a 3D DMLC tracking algorithm.

Zimmerman and colleagues tested their tracking system by applying two RapidArc plans - a lung plan and a prostate plan - to a target with simulated motion. Examining the dose delivered to this moving target with and without the use of the DMLC tracking algorithm revealed "a small, but consistent and reproducible, improvement in the dose distribution to the moving target" when the algorithm was employed. Zimmerman concluded that "a combination of RapidArc and the 3D DMLC tracking algorithm is feasible", noting that while this early feasibility test looks promising, further work is needed, in particular to test movement in more than one dimension.

Intensity-modulated arc therapies such as RapidArc and VMAT - along with other newer radiation delivery platforms like the TomoTherapy system and the CyberKnife - were themselves the subject of much discussion at ESTRO, with several conference sessions dedicated to these advanced treatment modalities. In a symposium examining the dosimetry of new beams, Karl-Axel Johansson from Sahlgrenska University Hospital in Sweden explained the issues associated with determining their absolute dose and relative dose distribution.

"These new beams all have small fields that result in a lack of charged particle equilibrium," Johansson explained. The problem is that existing dosimetry protocols require calibration measurements under reference conditions: usually a static beam with a 10x10 cm field size. But many of the newer beam-delivery modalities only have small field sizes and simply cannot produce these conditions. What's more, no recommendations exist for small beams. "We really need a totally new dosimetry protocol to cover these small beams," he added. "It's important that we have consistency in dosimetry when we change to these new beams."

Even if new protocols are created, it won't happen straight away and in the mean time, people still need to calibrate beams. "You can't wait for someone to write a nice protocol, you have to work on it yourself," commented Alan Nahum, from the UK's Clatterbridge Centre for Oncology. Nahum took a closer look at the dosimetry of small fields - such as those associated with stereotactic techniques or the tiny beam segments used with IMRT. In particular, he discussed the interpretation of detector response in small fields. The key issue here is detector size: small fields require small detectors, with the ionization chambers conventionally used for absolute dose determination too large to provide accurate results. "Measurements with different detectors will yield a spread of responses, due to averaging over detector volume," he explained. "The detector of choice has to be the Bragg-Gray cavity."

Elsewhere, there are plenty of other research groups looking to meet these new demands. Mauro Iori from the S. Maria Nuova Hospital in Italy presented results showing that electronic portal imaging device (EPID) systems - which are currently used for fixed-gantry IMRT dosimetry - can be extended for verification of intensity-modulated arc therapy plans. And Pascal Francois, from the Institut Curie in France, examined the issues of calibrating a TomoTherapy unit, where the typical calibration measurement conditions include a field size of 40x5 cm. Ultimately, each new treatment modality and every technology enhancement will come hand in hand with further hurdles to address. And inevitably, the physics community will be on hand to tackle and meet these emerging challenges.

Top picks
The five papers highlighted by this year's ESTRO Physics Committee are:

Motion-compensated cone-beam CT for accurate online assessment of the position of lung tumours
S. Rit et al. The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, the Netherlands.

A practical method to deal with differential motion of target and OAR during image guided radiotherapy
M. van Herk et al. The Netherlands Cancer Institute/Antoni van Leeuwenhoek Hospital, the Netherlands.

Introducing Pareto from evaluation as a concept for objective evaluation of inverse planning and delivery techniques
R. Ottosson et al. Copenhagen University Hospital, Denmark; Lund University Hospital, Sweden.

Quantifying the robustness of IMRT treatment plans from a statistical optimization model to account for organ deformation
B. Sobotta et al. Radiooncological Clinic, University of Tübingen, Germany.

DMLC motion tracking of moving targets for intensity modulated arc therapy treatment
J. Zimmerman et al. The Finsen Center, Rigshospitalet, Denmark; Varian Medical Systems (Palo Alto, CA); Stanford University (Stanford, CA).