First up, Michael Duchateau from UZ Brussel Oncology Centre in Belgium discussed some of the tools available for dosimetry of advanced dose distributions. He started by telling the assembled delegates some good news: that radiotherapy centres achieve a compliance rate of nearly 98%. As an example, he pointed out that when asked to deliver 1 Gy, 98% of institutions delivered 1 Gy.
He then pointed out the bad news: how in a recent audit, 30% of credentialed institutions failed to accurately deliver an intensity-modulated radiotherapy (IMRT) head-and-neck dose distribution (credentialing data courtesy of the Radiological Physics Center).
"New technologies are error prone, IMRT is still not plug-and-play," Duchateau explained. "It's a struggle for centres to reach the comfort zone and do better at the next credentialing."
Dosimetry selection
One newer option available for QA of three-dimensional dose distributions is an array – of either ionization chambers or diodes. Such arrays can have a linear design or (as introduced more recently) employ a rotational geometry. Linear arrays can be used for both field-by-field and composite dosimetry, and can perform machine QA. The downside is their significant angular dependence.
Rotational arrays – specifically designed for QA of arc therapies and tomotherapy, enable fast composite QA. Duchateau mentioned some of the newer commercial systems for rotational dosimetry, such as ArcCHECK (Sun Nuclear, Melbourne, FL) and Delta4 (ScandiDos, Sweden).
"Arrays offer a good alternative to film, you can do processing on the fly and lots of add-ons are available," he said. "The big point to remember is 'try before you buy', because lots of different arrays are available and you have to validate these before use in clinics. You have to QA the QA tools that you're going to use."
Another method employed for QA of advanced radiotherapies is dose reconstruction. This can be performed with built-in electronic portal imaging devices, where greyscale images are converted to dose, or by using add-on options such as COMPASS (IBA, Belgium) or Sun Nuclear's 3DVH.
Benefits of dose reconstruction include the fact that it is patient specific and users can define their own workflow. One downside is the large amount of data generated, which requires good ICT resources. Duchateau noted, however, that this method does not provide physical 3D dose measurement, only 3D reconstruction, and that the accuracy is based on a calibration or simulation model.
He also described the emergence of chemical gel dosimeters, based on monomers that polymerize upon exposure to radiation or radiochromic gels that change colour upon irradiation. "No other dosimeter can serve as a 3D dosimeter today," said Duchateau. "It's real 3D dosimetry that can be used for end-to-end testing."
The process of gel dosimetry is, however, relatively complex, involving the creation of a gel phantom, irradiation, and quantitative imaging (mostly via MRI or optical CT) of the resulting 3D dose distribution inside the gel. The overall accuracy is dependent upon the gel properties, as well as the read-out technique used. Research is now is focused upon improving usability, as well as minimizing toxicity.
In conclusion, Duchateau asked whether QA tools are keeping up with the rapid evolutions in treatment modalities. Noting that it's still difficult to reach the comfort zone for QA of techniques such as IMRT, he suggested that they are not.
Optical options
Following on from Duchateau's presentation, Claus Andersen, of Risø DTU at the Technical University of Denmark, told the ESTRO delegates about the use of online luminescence dosimetry for monitoring radiation therapy.
Luminescence dosimetry is performed using an optical fibre-based probe containing a small amount of organic plastic scintillator or inorganic crystalline phosphor. During irradiation, the scintillator spontaneously generates radioluminescence in proportion to the instantaneous dose rate.
As the millimetre-sized probe contains no electrical wires or other electronics, it can be used for in vivo dosimetry. The generated optical signal is recorded in real time via the optical cable, which can be connected to remote read-out devices.
As well as producing radioluminescence, some crystalline inorganic phosphors (such as carbon-doped aluminium oxide, Al2O3:C) also trap energy. After irradiation, this stored energy is released as luminescence by exposure to laser light, creating an optically stimulated luminescence signal that provides a measure of accumulated dose. After each use, the crystal can be "reset" via further laser treatment. Andersen notes, however, that this process can take five to 10 minutes, which could be problematic in a clinical setting.
Andersen also described some of the challenges associated with online luminescence dosimetry. Firstly, there's ionization quenching, which implies that the amount of light produced per absorbed dose unit is not the same in, for example, kV and MV X-ray beams.
The "stem effect", in which irradiation of the optical fibre itself produces light, also needs to be addressed. For inorganic phosphors, this can be achieved by exploiting the long lifetime of luminescence to separate it from the stem signal produced in pulsed linear accelerator beams. This approach cannot be used with the fast organic scintillators though, and other solutions such as chromatic separation are necessary. Andersen noted, however, that such plastic scintillators are of particular interest due to their direct water equivalence.
Promising applications for the probes include point measurements for in vivo dosimetry, as well as reference dosimetry in small fields. For brachytherapy dosimetry, for example, probes can be placed in unused channels of a standard brachytherapy applicator to perform in-tumour dosimetry. Another example is the use of a thin layer of Al2O3:C on the fibre for particle therapy dosimetry.
Andersen rounded up by emphasizing that online luminescence dosimetry is an emerging technology. When questioned as to why the method has not yet been commercialized, Andersen suggested that the use of delicate fibre cables within clinics may be an issue. "We need more demonstrations showing that it works in small studies and then we'll move on," he said.