medicalphysicsweb review

Transcription

1 medicalphysicsweb review News, analysis and opinion from medicalphysicsweb Summer 2009 How IMRT can become routine Why is IMRT not yet the standard radiotherapy modality? asks Marco Schwarz. Regardless of the specific event considered the starting point for intensity-modulated radiation therapy (IMRT), it s safe to say that the technique has been around for more than 10 years. How can it be, then, that in 2009 one writes an article examining how IMRT can become routine? The fact is, although some people may think that IMRT is fully mature and now represents the standard radiotherapy technique, this is not entirely the case. It makes sense, therefore, to try to identify the main causes underlying the delayed transition from three-dimensional conformal radiotherapy (3D-CRT) to IMRT. We should first remember that implementing IMRT following several years of 3D-CRT experience is a completely different challenge to carrying out the same development in a department where 2D radiotherapy is still the de facto standard. While this may be stating the obvious, we tend to forget one important reason why IMRT has not shown a faster adoption rate: when IMRT came into being, CRT (i.e., CT-based 3D treatment planning with conformal apertures, based on delineated contours for both the targets and organs-atrisk) was not the standard for every centre and/or treatment site, even in more affluent parts of the world. Furthermore, for clinicians in particular, it may be harder to move from 2D radiotherapy to CRT than from CRT to IMRT. If we take these aspects into account, the slow pace of IMRT implementation suddenly becomes easier to understand. Worlds apart From a technical perspective, the difficulty in transitioning from CRT to IMRT may be explained by one simple observation. Namely, that for quite some time, CRT and IMRT were considered by most clinics and by the companies producing radiotherapy equipment as two different worlds that could not be connected via a smooth path. Imagine, for example, that you want to treat a patient with prostate cancer and your choice is between a four-field CRT plan and an IMRT plan with 400 control points. Assuming that you are not comfortable with dose escalation, the two plans can in most cases be designed to deliver the same dose to the target, with IMRT allowing you to reduce the volume of rectum receiving higher doses. But in this situation, would you be excited enough by the IMRT plan to immediately transition all of Easy does it: further improvements in planning software, plus more efficient verification techniques, should help increase IMRT uptake. your prostate patients to IMRT? I am stretching things a bit, but I m sure that you see the point: in the early days of IMRT, it was extremely difficult to envision it as a logical extension of CRT. The radiotherapy community needed time to fully grasp the specifics of this new technique and to benefit from IMRT s additional degrees of freedom without making their lives unnecessarily complicated. In a way, people had to realize that even if you could deliver extremely complex segment patterns, this doesn t mean that you actually need to and/ or should do for every patient from day one given that this complexity usually comes at a cost. But what do we mean exactly by cost? If we think about it, it s not the hardware that s been the main obstacle to a widespread uptake of IMRT. The real cost lay in the complexity of the treatment-planning and verification process. In my opinion, improvements in treatment-planning software have played an essential role in making IMRT less cumbersome than it was 10 years ago. Getting used to the concepts that underlie treatment-plan optimization is, in itself, a fair obstacle in the transition from CRT to IMRT. If users of treatment-planning systems are faced with optimization software that makes it difficult to translate clinical requirements into a cost function and to steer the optimization towards an acceptable result, IMRT planning becomes a time consuming and frustrating task. If there s a single development in treatment planning worth mentioning here, it s the possibility of optimizing deliverable plans, as opposed to fluence distributions. The experience of the past years has taught us that this is a key step in improving the whole planning procedure. What is still missing from the optimization modules of most commercial treatment-planning systems is a means to gain insight into the optimization results; for example, the ability to identify which planning objectives are most important in determining a given solution and to evaluate which adjustment to the cost functions could produce better results. When available, such tools will allow the planner to better trust the final plan, which will be not just a plan, but a dose distribution carefully chosen In the early days of IMRT, it was difficult to envision it as a logical extension of CRT. from several realistic alternatives. The increased complexity in segment patterns that characterizes IMRT was not matched, particularly in the early years, by an adequate increase in the accuracy of dose calculation. With CRT, tasks such as MLC quality assurance and treatment-planning-system commissioning were strictly decoupled from the evaluation of the dosimetric accuracy for a single treatment plan. With IMRT, however, this separation mostly disappeared. Pre-treatment dose verification of each treatment plan thus became a routine part of the IMRT process a clear indicator that the accuracy of the dose calculation was not fully trusted and that it was difficult to define a priori accuracy criteria for the planning system that were achievable and would provide full confidence in every single plan. In many clinics, the time required to carry out such pretreatment dose verification is still the real bottleneck hindering large-scale IMRT implementation. There are two possible solutions to this problem: make the verification process fast and efficient, or somehow avoid the need for it altogether. With regard to the first approach, the possibility of combining all treatment-verification steps, by performing EPID-based in vivo dosimetry during the first treatment session(s) is particularly interesting. Others argue that there is no reason why the accuracy criteria and approaches deemed valid for CRT should not apply to IMRT too, and that dosimetric pre-treatment verification should be avoided by implementing better dose algorithms in commercial treatment-planning systems. In practice, this means that Monte Carlo dose calculations should become standard. The large-scale availability of Monte Carlo-based dose calculations in clinical practice has been anticipated for years, but remained largely an unkept promise. One of the causes for this delay, was actually the interest in IMRT, which further shifted the focus from accuracy to speed in dose calculation. We are now at a time where high accuracy and satisfactory speed can be combined. Thus there is no reason why Monte Carlo should not become the standard dose engine for IMRT planning in the near future. Marco Schwarz is a physicist at the Agenzia Provinciale per la Protonterapia in Trento, Italy. EDITORIAL Welcome to medicalphysicsweb review, a special supplement brought to you by the editors of medicalphysicsweb. This issue, distributed exclusively at the AAPM Annual Meeting, brings you a selection of our recent online content including special focuses on radiotherapy, proton therapy, nuclear medicine, diagnostic imaging and nanomedicine. If you ve not yet come across medicalphysicsweb, this supplement should give you a taster of what the site is all about. If you like what you see, visit the website to read more in-depth research articles and check out the latest features, such as videos and our recent webinar. For full access to all premium-content articles, why not sign up as a member of medicalphysicsweb. Membership is free and also offers you a weekly newswire delivered direct to your desk, the ability to comment on breaking stories and free corporate listing in our Buyer s Guide. To join our community, simply visit medicalphysicsweb.org and hit sign-up now. Alternatively, come and visit us at booth #911, where you can also find out about our exclusive AAPM prize draw to win an ipodtouch. Tami Freeman Editor, medicalphysicsweb SIGN UP TODAY Register FREE as a member of medicalphysicsweb at AAPM and you ll be entered into our fantastic prize draw for the new ipod Touch. Simply complete the registration form on p 12 and return it to our staff at booth #911. The winner will be announced Wednesday morning. Sign up as a member at medicalphysicsweb.org medicalphysicsweb review Summer 2009

2 ANATOM-e XRT New OEM Relationship New Academic Site License Program for Med Schools Super New Super Useful Anatom-e Information Systems, Ltd. Live Instant Way to Train. See. Consult. The Industry s New Knowledge Piece Imagine being able to quickly access the very best information about medical physics, biology, nano tech and more... Red Journal to Debut in ASTRO Pilot Speed-Contouring Call Now: Ask about Anatom-e U Ask about our contouring service bureau Hot! AAPM Edition Anatom-e s Brain Product Voted The Very Best. It s About TIME Anatom-e provided the base educational platform for the ARS Post Graduate Course 2009 Ask about obtaining CME

3 focus on: radiotherapy 3 MRI-accelerator: the proof of concept Incorporating real-time image guidance into radiotherapy would ramp tumour targeting accuracy, enabling better avoidance of critical structures and reducing side effects. Such guidance will be of particular benefit if a non-ionizing imaging technique such as MRI is employed. As such, a research team at the University Medical Center Utrecht in the Netherlands is working to integrate a linear accelerator with an MR scanner. Now the Utrecht team has demonstrated proof-of-concept operation of their system (Phys. Med. Biol. 54 N229). In a step claimed to open the door to start testing MRI-guided radiation therapy in the clinic, the team showed that MR imaging with the radiation beam switched on does not degrade the performance of either the linac or the MR scanner. The key significance of this work is the fact that we show that real simultaneous irradiation and diagnostic-quality MR imaging is feasible, said Bas Raaymakers, from UMC Utrecht s department of radiotherapy. Before implementation in the radiotherapy clinic we obviously need to make a few more steps, but the proof is there that makes it worth investing in these steps. Design features The prototype device comprises a 6 MV linear accelerator (Elekta, UK) positioned laterally to a 1.5 T Achieva MRI system (Philips, Netherlands), with a source-to-isocentre distance of 1.5 m. Ultimately, the accelerator will be mounted on a ring gantry. The researchers modified both the MRI and accelerator to enable their simultaneous and unhampered operation. The linac was customized by replacing various steel components with non-ferromagnetic versions, as well as mounting it on a wooden frame (instead of its steel gantry), while the MR scanner was equipped with a replacement magnet and gradient coil. The MR magnet was built (by Magnex, UK) with the central 15 cm free of coils to let the radiation beam through, giving a maximum irradiation field of 24 cm in the head-feet direction. The magnet is actively shielded, with most of the external field generated by the inner coils cancelled by a field generated from a pair of shield coils. The configuration is tuned to provide a toroidal low-field zone around the magnet. The most sensitive parts of the accelerator are then sited in this low-field region. The gradient coil, meanwhile, (designed by Philips Research, Hamburg, Germany) has a central 19 cm free from copper windings and offers imaging performance comparable to that of a standard Achieva coil. Raaymakers explained that system development posed a variety of challenges on all levels. The magnet did not fit through the door so we had to break open our radiation bunker, and we got some manuals in Chinese, he said. But the scientific breakthrough was this concept of active shielding, which magnetically decoupled the accelerator and MRI to a large extent. Finally, the team also redesigned the treatment room set-up. Instead of using the standard RF shielding method placing the MRI inside a Faraday cage shielding was achieved via two RF cages situated at either side of the MRI bore. In this design, the inner wall of the MRI cryostat becomes an integral part of the RF cage and the sample volume is shielded from the rest of the room, including the accelerator. Together, the magnetic decoupling and the RF decoupling made it possible to perform MRI with the radiation beam on, Raaymakers explained. Proof of concept The researchers performed initial imaging tests on volunteers (with the accelerator switched off) using standard sequences for prostate, brain and kidney MRI. All images were of diagnostic quality. They then examined the simultaneous use of the MRI and accelerator systems, by performing 1.5 T MR imaging on a pork sample during irradiation. Images taken with the radiation beam on and off were identical, and no degradation of linac performance was seen. Working towards a clinical prototype, the next step is to incorporate a multileaf collimator, which needs to be non-magnetic. Meanwhile, constructing a gantry for accelerator rotation will facilitate treatments using an arc-therapy approach. The Utrecht team is currently working with Elekta and Philips on these tasks. We hope to start the Tumour tracking, with sub-mm precision Jan Kok Ultimate integration: Raaymakers works on the installation of the Utrecht team s prototype MRI-linac. Advances in technology over the past decade have made it possible to image, plan and deliver radiation to a three-dimensional volume with sub-millimetre precision. So long as the target tumour is stationary, that is. Precision planning counts for nothing if the object of interest keeps moving in and out of line. What s needed is a delivery system that responds rapidly to tumour motion and reorients the treatment beam accordingly. Researchers from Stanford Cancer Center (Stanford, CA) and Washington University School of Medicine (St Louis, MO) have now demonstrated that this is possible by combining electromagnetic localization technology with real-time, dynamic multileaf collimator (DMLC) tracking (Int. J. Radiat. Oncol. Biol. Phys ). The work was carried out in collaboration with Varian Medical Systems (Palo Alto, CA) and Calypso Medical Technologies (Seattle, WA). Calypso s 3D position monitoring system, which uses an electromagnetic planar array to excite and receive signals from three wireless transponders, was altered so that transponder positions were updated at a frequency of 25 Hz instead of 10 Hz. This increase in update frequency was needed so that respiratory motion could be followed. Dynamic tracking software based on a 120-leaf MLC unit, and developed previously by the academic researchers, was also modified to read real-time target position from the Calypso system. The researchers assessed the time lag between target motion and MLC response using a continually-moving phantom fitted with three electromagnetic responders. The phantom s movements were simultaneously tracked by a circular MLC field and recorded on megavoltage X-ray imaging (around 7 frames per second). The geometric accuracy of the DMLC tracking set-up was investigated separately using two respiratory traces (one with high variability, one with moderate variability) and a trace showing highly variable prostate motion. Each trace was programmed into the motion platform holding the phantom. Real-time MLC tracking and X-ray imaging were carried out as before. These tests revealed the system s latency to be approximately 220 ms, whilst the tracking accuracy was sub-2 mm for the respiratory motion traces and sub-1 mm for the prostate motion. Sub-2 mm is regarded as clinically acceptable accuracy, although we are currently working on achieving sub-1 mm, lead author Amit Sawant, instructor in radiation physics at Stanford, told medicalphysicsweb. In the absence of real-time tracking, practitioners must account for tumour motion by drawing significant margins around the target, Sawant explained. For large tumours that are mobile, the volume of normal tissue covered in the safety margin can be considerable. Gating methods allow margins to be reduced, but this option is not ideal. Gating reduces the duty cycle, which means longer treatment time. This in turn means a higher probability of motion, a higher probability of cell recovery and reduced patient throughput, Sawant said. One drawback of the electromagnetic positioning system is the need to implant transponders in or around the tumour prior to radiation treatment. Work is under way to find the best method of doing this safely in the lung, where the risk of organ puncture is a real concern. The possible migration of transponders over the course of therapy requires further investigation too. We are currently working on a comprehensive evaluation of our system over a variety of tumour motions from several sites (lung, liver, prostate), along with several delivery techniques (conformal, IMRT, volumetric modulated arcs), Sawant said. We are also working on developing QA procedures for this system. Paula Gould is a contributing editor on medicalphysicsweb first clinical tests in a year s time. Whether this involves patients remains to be seen, Raaymakers told medicalphysicsweb. He continued: We are currently discussing with our physicians what the best introduction scheme is. Should we start with palliative patients, start by using MRI as improved position verification or should we start with a novel strategy such as irradiation of liver metastases? This choice will determine the technical requirements such as, for instance, full gantry rotation, and the accompanying time schedule. Tami Freeman is editor of medicalphysicsweb Upping the dose via tomotherapy Preliminary results from an ongoing clinical trial reveal that helical tomotherapy may enable higher biologically-effective radiation doses to be delivered to non-small cell lungcancer patients, with lower than expected toxicities. In contrast with most dose escalation studies, the researchers from the University of Wisconsin School of Medicine and Public Health (Madison, WI) did not increase the total number of fractions. Instead, all patients received 25 treatment fractions over five weeks, with the dose set according to each individual s likelihood of lung toxicity (TCRT 6 441). For the 46 patients in the study, the overall two-year survival rate was 46.8%, a large increase over historical rates of 21.5% for the same stagerange of disease. The study showed that higher doses of radiation (typically around 60 Gy in 2 Gy fractions) could be delivered safely using this hypofractionated tomotherapy schedule. Lung and oesophageal toxicities were lower than expected. Sign up as a member at medicalphysicsweb.org medicalphysicsweb review Summer 2009

4

5 focus on: radiotherapy 5 Dosimeter targets small fields A Patents round up of the latest international patent applications in radiotherapy. Brachytherapy: balloon options Lumpectomy has become a standard treatment for women with earlystage breast cancer. Survival rates for this less radical surgery are equivalent to those of mastectomy, so long as the procedure is followed by whole-breast external-beam radiation therapy (EBRT). Therein lies the problem. Postoperative radiotherapy can take several weeks to complete. Guaranteeing attendance is difficult, particularly if patients have to travel far. But if the course of radiotherapy is not completed, or worse, not even started, then there is a greater risk of the tumour reappearing. An obvious solution to this problem is to speed up the radiotherapy regimen. This can be done by reducing the volume irradiated to just the lumpectomy bed, and increasing the size of each fraction. Follow-on radiotherapy can be finished within a week with this more focused approach. Several centres have trialled multicatheter interstitial brachytherapy as a way of delivering radiotherapy in this manner. Results indicate that this technique is just as effective as whole-breast EBRT for combating tumour recurrence. However, this method can be fiddly and requires considerable experience. Another option for breast brachytherapy is to use a single, doublelumen catheter attached to an inflatable balloon. One such device is the MammoSite from Cytyc (Marlborough, MA, a Hologic company). DOSI prototype: the 128-channel dosimeter with its electronics. High-resolution dosimetry is important for all radiation therapy techniques, but most critical for the newer conformal treatment modalities that employ small radiation fields. At the extreme, stereotactic radiotherapy and radiosurgery typically use field sizes of less than 4 cm, and at times as small as 5 mm. Such regimes necessitate tighter target margins and pose stringent dosimetry demands. There are, however, some big challenges associated with small-field dosimetry. First, as field size decreases, the dose becomes inhomogenous across the beam. Increased collimation also results in steeper dose gradients at the beam edges. If detectors larger than the beam s penumbra are used to measure such fields, they systematically overestimate the beam width errors that could have significant clinical impact. What s needed is a detector with an active width smaller than the beam dimensions that can perform stereotactic beam dosimetry with high spatial resolution and in real time requirements that are currently not met by any commercially available product. To fill this gap, UK researchers are developing a smallfield dosimeter based on positionsensitive detectors that were originally employed in particle physics (Phys. Med. Biol ). The device, called DOSI, is capable of real-time measurements with submillimetre resolution. The research team from the Rutherford Appleton Laboratory (RAL), University Hospital Birmingham NHS Trust, the University of Birmingham and Swansea University investigated a prototype DOSI made from monolithic silicon implanted with a linear array of 128 diodes (developed by an earlier RAL collaboration). The balloon is inserted into the lumpectomy cavity and then inflated with a mixture of saline solution and radiographic contrast material, allowing its position to be verified on imaging. The prescription dose is then delivered in multiple fractions through the catheter, again using a radioisotope source. A review of available data has now given this form of balloon brachytherapy a cautious thumbs up. Initial reports suggest that the userfriendly method is comparable to both multicatheter-based interstitial brachytherapy and whole-breast EBRT in terms of patient outcome and cosmetic advantage (Radiother. Oncol. doi: /j.radonc ). More evidence is needed, though, before any definitive recommendations can be made, according to the report s authors. Emerging methods of balloon brachytherapy may improve access to partial-breast irradiation further. For example, Xoft has developed a balloon-compatible, disposable brachytherapy source that works with kv X-rays rather than radioisotopes. Another up-and-coming development is Contura, a multilumen balloon applicator from SenoRx (Irvine, CA). The study authors recommend that patients who receive this type of accelerated radiotherapy are enrolled in trials, and that their progress is monitored for a significant period of time. Careful long-term follow-up is important for the welfare of individual patients and to establish the efficacy, late toxicity and cosmetic outcomes of this technique, they concluded. The team used the prototype device to characterize 6 MV photon beams from a stereotactic collimator system, with beam diameters ranging from 7.5 to 35 mm. The device s performance was evaluated by measuring basic beam data, namely, percentage depth dose distribution (PDD), off-axis ratio (OAR) and relative output factor (ROF). The DOSI data were compared with corresponding results from the hospital s standard dosimeters: a diamond detector and a PinPoint ionization chamber, both optimized for stereotactic measurement. Excellent Polymers eye 3D dosimetry Researchers at the Technical University of Delft in the Netherlands are developing a new method of radiation dosimetry: radio-fluorogenic co-polymerization (RFCP), which is based on radiation-induced polymerization of a liquid monomer (Phys. Med. Biol ). We were motivated by the idea that the RFCP effect could ultimately be used to make 2D and even 3D fluorescence images of the dose distribution in tissue-equivalent phantoms for testing radiological treatment procedures, explained John Warman of the university s Reactor Institute Delft. If the polymer could be formed within a quasi-rigid gel matrix, then it would be immobilized and a fixed fluorescence image of the spatial distribution of radiation dose would be created. The team examined two potential RFCP solutions: the fluorogenic compound N-(1-pyrenyl) maleimide (MPy), mixed with either methyl methacrylate (MMA) or tertiarybutyl acrylate (tbua). To determine dosimetry performance, they recorded the fluorescence output of each solution while it was moved into and out of the irradiation chamber of a cobalt-60 gamma-ray source. For both probes, the fluorescence output increased linearly with accumulated dose. When the chamber was raised, the output intensity remained near-constant, indicating minimal post-irradiation effects. The higher sensitivity of tbua enabled data with correlation was seen between DOSI s results and those recorded with the diamond detector, with PDDs in agreement to better than 1% for all depths. The agreement between OARs was better than 3% for all collimators, while the agreement on ROFs was at the 1% level. For smaller beam diameters, particularly those of less than 15 mm, there were noticeable differences between the beam profiles and ROFs measured with DOSI and the Pin- Point ionization chamber. This is attributed to the volume averaging effects experienced by large aircavity ionization detectors. The diodes in DOSI are 0.25 mm wide, Manolopoulos told medicalphysicsweb. This is much smaller than the diamond, which is 1 mm diameter at best, and 20 times smaller than the 5 mm diameter Pin- Point. Manolopoulos points out that DOSI is also orders of magnitude faster than its rivals. Looking to the future, Manolopoulos envisages DOSI being used with advanced radiotherapy techniques. This is already taking place at UHB, where we re about to start IMRT treatment. Gel dosimetry: a radio-fluorogenic gel fluoresces after irradiation by a masked 3 MV electron beam. a reasonable signal-to-noise ratio to be obtained using an integration time of just 2 s. Comparing the performances of the two solutions in a radiotherapy-relevant regime (0.27 Gy/ min with an accumulated dose of 2.5 Gy) revealed that only tbua/mpy was accurate in this range. The researchers concluded that RFCP s linear dose dependence, fast response time and large dynamic range make it suitable for real-time radiation dosimetry. Meanwhile, the possibility of creating a polymer gel could ultimately enable 3D dose imaging during radiotherapy. We have been able to produce an optically clear radio-fluorogenic gel matrix in which the spatial resolution of the fluorescence image is 0.1 mm or better, Warman said. We hope, in the not too distant future, to demonstrate that the method can be used to image multi-directional X-ray beams, proton-beam tracks and the dose distribution around radioisotope sources used in brachytherapy. MLC enables real-time response Elekta of Sweden has developed a radiotherapy system that enables real-time adjustment of beam shape in response to, for example, online CT scanning of a patient during treatment (WO/2009/056151). The apparatus includes a therapeutic radiation source and a multileaf collimator (MLC) for delimiting the radiation beam. The collimator leaves can move longitudinally to define a variable collimator edge, and are mounted on a support that can move laterally with respect to the leaves. This set-up enables tumour movement perpendicular to the direction of leaf motion to be accommodated by simply moving the collimator bodily. Therapy minimizes collateral dose Orbital Therapy of Bedford, MA, has designed a radiotherapy system for treating extremities, in particular the breast, with minimal collateral dose to the rest of the patient (WO/2009/022197). The system includes a positioning table, on which the patient lies in a prone position while the radiation source rotates about the patient. This geometry maximizes the separation between the target volume and adjacent critical structures such as the chest wall, heart and lungs. The table is made from a shielding material to reduce or eliminate unwanted patient dose from scattered or stray radiation. A shielded enclosure for the radiation sources, combined with the shielded interface surface, eliminates the need for primary shielding in the treatment room. Mini source targets skin cancers A miniature X-ray source for radiation therapy of skin cancers is described by Nucletron of the Netherlands (WO/2008/155092). The source comprises: a vacuum tube containing a cathode and an anode spaced at some distance from each other; a means for emitting free electrons from the cathode; an electric-field generator for applying a high-voltage field between the cathode and the anode (in order to accelerate free electrons towards the anode); and an exit window for the X-ray radiation generated when the electrons impact on the anode. Applicator shields sensitive tissue Brachytherapy specialist Xoft (Sunnyvale, CA) has come up with a scheme for modifying the distance from a brachytherapy source to sensitive anatomical structures (WO/2009/017707). The balloon brachytherapy applicator contains one or two chambers with the ability to blister outwards at a specified area on one side of the balloon thus moving sensitive tissue away from the radiation source inside the balloon. In one form, a secondary compartment is formed on the primary balloon that s separately inflatable to allow blistering as needed. Sign up as a member at medicalphysicsweb.org medicalphysicsweb review Summer 2009

6 6 focus on: proton therapy Protons decrease second-cancer risks Advances in cancer detection and treatment have led to large improvements in survival rates. But with increased survival comes an increased need to minimize longterm treatment-related effects. In particular, children receiving radiation therapy are more susceptible to radiation carcinogenesis and have an increased risk of developing second cancers later in life. Proton therapy offers a theoretical dosimetric advantage over photon radiotherapy, in terms of sparing nearby tissues and organs. As such, it has been proposed as a promising modality for treating central nervous system cancers in paediatric patients. But, while several studies have examined the stray radiation exposures associated with proton therapy, the risks from neutrons generated during such treatments have, in the main, been neglected. With this in mind, researchers at the MD Anderson Cancer Center (Houston, TX) have compared the risks of developing a second cancer following craniospinal irradiation (CSI) delivered via four modalities: conventional photon therapy, intensity-modulated photon radiotherapy (IMRT), passively scattered proton therapy, and scanned-beam proton therapy, with neutron generation accounted for in the proton treatments (Phys. Med. Biol ). To reduce unwanted radiation dose to healthy tissues, proton medicalphysicsweb review Summer 2009 beams have been used to deliver CSI to children with medulloblastoma, explained Wayne Newhauser, from MD Anderson s department of radiation physics. However, proton therapy also produces stray neutron radiation, which is approximately a factor of 10 more carcinogenic than photon radiation. Prior to our study, neutron exposures caused by proton CSI were not known. Newhauser points out that this knowledge gap hinders evidencebased decisions on treatment-modality selection and healthcare policy. He also notes that previous publications speculating on the possible dangers of proton therapy have caused confusion as to the appropriateness of proton therapy for treating children whereas, in fact, evidence now shows that the risk is lower following proton therapy, regardless of whether the proton beams are produced with passive-scattering systems or magnetically scanned beams. Risk assessment The MD Anderson team simulated a paediatric medulloblastoma treatment based on CSI with a prescribed absorbed dose to the target volume of 36 Gy. Monte Carlo code was used to calculate the stray radiation doses from neutrons for passively scattered and scanned-beam proton treatments of a male phantom. The organ doses were then converted to risk of cancer incidence. This lifetime risk Big step for small synchrotron The emergence of lower-cost protontherapy systems based on compact accelerators will enable many more patients to benefit from this highly conformal cancer treatment. To this end, several commercial firms are developing compact systems based on a range of technology approaches. One such company is ProTom International (Flower Mound, TX), which teamed up with MIT s Bates Linear Accelerator Center (Middleton, MA) last summer to validate its compact proton-therapy system. Now, the partners have announced successful beam acceleration and extraction from the ProTom system. The ProTom synchrotron arrived at MIT-Bates earlier this year for validation testing. Following just two months of installation and commissioning, a proton beam was extracted with a beam energy ranging from 30 to 250 MeV. The research team also demonstrated dynamic real-time modulation of the beam intensity within each extraction cycle. These tests, overseen by MIT engineers and scientists, represent the first successful demonstration of a proton beam of clinical range from a small-footprint system that will be commercially available in the US, said ProTom s CEO Stephen Spotts. We now look forward to completing the formal FDA [Food and Drug Administration] review process in order to make this important new treatment tool clinically available to cancer patients in the US. The proton-therapy system is based on next-generation synchrotron technology originating at the Lebedev Physics Institute in Russia, and to which ProTom has secured exclusive US commercialization rights. The synchrotron has an external-ring diameter of around 16 feet (4.9 m) and weighs around 15 tonnes. A range of technical advances facilitate proton-beam scanning with a variable pencil-beam size. What s been achieved with our device should be considered as a design breakthrough, explained Pro- Tom director Andrew Cowen. It s a conventional synchrotron, but with many enhancements and improvements - for example, in terms of magnet design, ring construction and the extraction process - that enable us to significantly scale down the size and dramatically reduce the number of components. It s certainly a compact and streamlined system, but it s in a very different category relative to the Stray dose: the absorbed dose for the entire CSI and boost treatment superimposed on the patient s CT images. The coronal plane is shown for (a) external neutrons, (b) internal neutrons and (c) all neutrons. devices being developed by Still River Systems and TomoTherapy, which I would characterize as more in the area of technology breakthrough. The ProTom system has already demonstrated pulse-to-pulse energy variability. It also has the capability to perform dynamic energy modulation during a pulse. At this point, I believe our system is the only one specifically designed for, and capable of achieving, proton-beam scanning with dynamic energy and intensity modulation in three dimensions, Cowen told medicalphysicsweb. The MIT-Bates researchers will continue to work with ProTom to provide independent validation of the system and help identify further potential enhancements. They are also helping ProTom generate data for the FDA 510k approval necessary to begin treating patients, which Cowen predicts should be in place by the end of this year. The exciting point for ProTom is that we ve actually got a working machine, which I think puts us several steps ahead, said Cowen. The market benefits from having competition and alternatives; but now we can demonstrate our capabilities, it s a big step for us and for the field. mgy of developing a second cancer due to stray radiation was 1.5% for passively scattered proton therapy and 0.8% for the scanned-beam treatment. The researchers next considered the added risk associated with exposure to the primary therapeutic proton beam, and came up with an estimate for the total second-cancer risk of proton therapy. They note that the risks associated with proton therapy are predominated by primary radiation and not stray neutron radiation. These data were compared with the second-cancer risks associated with conventional 6 MV photon radiotherapy and 6 MV IMRT. Results reveal that proton therapy offers substantially lower risk than photon therapy, even when neutrons are taken into account. The total predicted lifetime risks of developing a second cancer following IMRT and photon radiotherapy were 31 and 55%, respectively. These values are seven and 12 times higher than that of scanned proton therapy (4.4%), and six and 11 times higher than the second-cancer risk associated with passively scattered protons (5.1%). Protons focus on prostate cancer In a separate study, Newhauser and colleagues examined the stray radiation dose and associated second-cancer risk for a 10-year-old boy who received CSI using passively scattered protons. Here, the Monte Carlo calculations were based on simulation of a proton CSI treatment plan and whole-body kilovoltage CT images from an actual paediatric patient treated at MD Anderson (Phys. Med. Biol ). This represents an incremental increase in the realism of our risk assessments, Newhauser noted. The proton-therapy plan created for the patient prescribed CSI at 30.6 Gy, plus a boost of 23.4 Gy to the clinical target volume. The effective dose from stray radiation for the entire treatment was calculated as 418 msv roughly equivalent to the effective dose from 20 whole-body CT scans. The stray radiation comprised 82% from external neutrons (originating in the treatment unit) and 19% from neutrons originating within the patient. The researchers estimated the corresponding excess lifetime risk of second-cancer fatality from stray radiation for paediatric male patients who receive this treatment at 3.4%. They note that while this risk is small compared with the benefits of the radiotherapy, it is not negligible. As such, it is important to continue attempts to reduce stray radiation exposure as much as possible. Long-term survivors of prostate cancer who were treated with radiation therapy face a small but significant risk of developing radiation-induced secondary cancer. Many of these second cancers arise in the normal tissue adjacent to the target, thus it s been suggested that proton therapy with its dosimetric advantages could potentially reduce their incidence. Studies have suggested that using spot-scanned proton therapy for prostate irradiation can reduce the incidence of second cancers compared with intensity-modulated radiation therapy (IMRT). However, as most clinical centres currently use passively scattered proton delivery, for which the production of stray neutron radiation has become a concern, it s important that this option is also investigated. Researchers at the MD Anderson Cancer Center (Houston, TX) assessed the relative risks of treating prostate cancer with passively scattered protons and 6 MV IMRT. The studies accounted for contributions from both primary and secondary sources of radiation (Int. J. Radiat. Biol. Oncol. Phys ). To compare the two techniques, the researchers used a commercial treatment-planning system to construct proton therapy and IMRT plans for three patients with early-stage adenocarcinoma of the prostate. The plans were used to calculate the absorbed doses from the primary field in adjacent organs at risk. Secondary doses were determined using Monte Carlo simulations for proton therapy and measured data for IMRT. Both treatments provided adequate coverage of the target volume and margins, with the proton therapy plan offering slightly greater dose uniformity. Both also provided acceptably low doses to surrounding critical structures. However, proton therapy delivered substantially lower doses at low and intermediate levels in the bladder and rectum. Having determined the excess relative risks of developing a radiationinduced cancer after proton therapy or IMRT, the researchers calculated the ratio of excess risk for proton therapy relative to IMRT. This ratio was 0.61, 0.66 and 0.74 for the three patients, suggesting that the proton therapy plan could reduce the second-cancer risk by 26 39% compared with IMRT. The MD Anderson team concluded that, despite the secondary neutrons, passively scattered proton therapy reduces the projected incidence of radiogenic second cancers in prostate-cancer patients. Sign up as a member at medicalphysicsweb.org

7 focus on: proton therapy 7 Lower energies: reduced costs There s no doubt that proton therapy has a great deal to offer clinically: its ability to conform the deposited dose to the intended target improves normal-tissue sparing compared with conventional X-ray therapy. However, with current estimates putting proton-therapy costs at 2.4 times those of intensity-modulated radiation therapy (IMRT), opponents claim that its benefits do not outweigh the additional expenditure. To redress this drawback, research teams are developing lower-cost proton-therapy systems, using compact accelerators that can be mounted on a rotating gantry and contained within the treatment room. Several approaches are under investigation, including miniaturized synchrocyclotrons, laser-induced plasma wakefield accelerators, and the dielectric wall accelerator (DWA), which uses pulsed electric fields generated inside a high-gradient insulating acceleration tube to speed protons passing through to high energies. A key factor when designing any new accelerator is specifying the maximum proton kinetic energy that the treatment system must achieve. While the maximum kinetic energy required is generally accepted as 250 MeV (giving a proton range of about 38 cm in water), recent research has revealed, actually, that Team work: the research groups of Paul DeLuca and Rock Mackie are studying many aspects of the DWA proton-therapy system. a lower value may be sufficient for the majority of proton treatments. Researchers from the University of Wisconsin School of Medicine and Public Health (Madison, WI) have quantified the maximum proton energy necessary to treat a given percentage of patients, and examined the implications of this energy threshold when building a gantrymounted proton-accelerator treatment system (Med. Phys ). The team analysed 100 randomly selected treatment plans from patients treated with IMRT for cancers of the breast (12), brain, head and neck (28), lung and bronchus (20), abdomen and pelvis (20) and prostate (20). For each case, they calculated the maximum proton-beam energy required to treat the patient using a DWA-based system. Calculations revealed that treating all patients in the study would need proton energies of up to 240 MeV. However, 90% of patients could be treated at 198 MeV, and 95% at 207 MeV. A beam kinetic energy of 200 MeV immediately prior to entry into the patient was sufficient to treat all breast, brain, head and neck cases, as well as 19/20 of lung and bronchus, 15/20 of abdominal and pelvis, and 17/20 of prostate patients. Decreasing the maximum kinetic energy from 250 to 200 MeV will decrease the length of the accelerator. This could lead to a significant reduction in the total volume of the treatment system and the size of the treatment room needed to house it, explained Evan Sengbusch, graduate research assistant in the University of Wisconsin s medical physics department. Secondly, accelerating protons to lower energies means that a lower maximum magnet strength is required to deflect and/or bend the proton beam. Even with a maximum kinetic energy of 200 MeV, Sengbusch predicts that a DWA proton system will, for now, likely still cost more than existing IMRT set-ups, although considerably less than current proton-treatment systems. The researchers conclude that it s feasible to significantly lower the maximum kinetic-energy requirements of a compact proton accelerator if the ability to treat a small percentage of patients with rotational therapy is sacrificed. Such a decrease could reduce costs and ease engineering constraints when setting up such a facility, allowing more patients access to this promising treatment modality. Robotic system speeds set-up IBA of Belgium has received FDA clearance for a robotic patient positioning system (PPS) for patients undergoing proton therapy. The PPS was developed by IBA and ProCure Treatment Centers (Bloomington, IN), in collaboration with Forte Automation Systems (Rockford, IL). This new robotic PPS is a costeffective, custom designed medical robot under computer control that decreases the time needed to position patients for proton treatments, said Niek Schreuder, senior vicepresident of medical physics and technology at ProCure, and principal designer of the device. The PPS has the advantages of being more userfriendly and flexible, with a greater freedom of motion than currently available systems. The basic configuration of the robotic PPS is a Selective Compliant Assembly Robot Arm (SCARA). It combines commercially available, cost-effective robotic subsystems with custom-designed arms that allow for the specific needs of patient positioning in proton therapy. A unique dual coupler system at the end of the wrist of the robot arm can be used to attach different devices to the PPS, such as a patient table or chair, or a test phantom. The device employs safe robotic technology that meets the standards of the Robotics Industry Association (RIA). Radcal s My Choice! ACCU-PRO Multi-Purpose - diagnostic x-ray analyzer For: Radiography, Fluoroscopy, Mammography, CT, Dental, Scatter/Leakage Booth 200 Measurement Ion Chambers, Dose Diodes kv Sensors, ma/mas Sensors RAPIDOSE SOLUTIONS - FOR ALL YOUR X-RAY MEASUREMENT NEEDS! The Gold Standard Ion Chamber System and the New Revolutionary Diode Detector X-ray analyzer Rapid Measurements - with a super small footprint detector For: Radiography, Fluoroscopy, Dental, Mammography (available soon) More Productivity Rapid Measurements Laptop not included Simultaneous Measurements Customizable User Interfaces Radcal 426 West Duarte Road Monrovia, CA USA T: (626) F: (626) in Radiation Gold Standard The

8 Look how far multi-lumen balloon brachytherapy has come. The pioneer and now the PROVEN LEADER in MLB technology. 400 sites, over 4,500 cases and rising. Radiation Oncologists, Medical Physicists and Breast Surgeons agree the rising star of breast brachytherapy is Contura MLB. The promise of Contura is now being clinically confirmed. With its proprietary, multi-lumen dose shaping capability, Contura enables you to optimize dosimetric coverage to the targeted tissue while minimizing radiation to healthy tissue. Contura s optimal design, proven clinical performance coupled with exceptional physics training programs have made it the first choice in dose shaping technologies. If you haven t done so already, there simply has never been a better time to adopt this advanced technology into your practice. Even though this patient presented with very tight skin and rib margins, treatment with Contura MLB was successful due to its ability to precisely shape the radiation dose.

9 OPTIMAL design Multi-lumen dose shaping optimizes treatment coverage Vacuum assist facilitates improved target coverage Exceptional balloon symmetry PROVEN performance Minimize skin and rib dose Flexibility to treat more patients More clinical data on the advantages of MLB EXCEPTIONAL support A choice of training programs makes it easy to get started Customer Service customerservice@senorx.com Rx Only. Copyright 2009, SenoRx, Inc. All rights reserved. SenoRx, Inc., Irvine, CA MM0204B

10 10 focus on: nuclear medicine PET: predicting drug response Chemotherapy plays a vital role in the treatment of cancers, but its therapeutic success is not guaranteed. A patient may respond to one drug but not another, or a tumour may mutate and stop responding to a drug. The ability to predict the effectiveness of a particular drug could prevent unnecessary courses of chemotherapy and the associated toxic side effects. PET scanning could provide such a predictor. Researchers at the University of California, Los Angeles (UCLA) are developing a PET-based tool that could one day allow doctors to evaluate an individual tumour s response to a chemotherapy drug before prescribing therapy, enabling them to personalize treatment to a patient s unique biochemistry. The UCLA research team has previously created a PET probe by slightly altering the molecular structure of gemcitabine, one of the most common chemotherapy drugs (Nature Med ). The probe, called 18 F-FAC, is labelled with positron-emitting particles that allow its movement through the body to be tracked via PET imaging. In this latest study, the researchers injected the probe into mice with leukaemia and lymphoma tumours, and then imaged the animals an hour later. Gemcitabine is activated intracellularly at a rate controlled by an enzyme called DCK, the activity of which varies significantly among individuals and across different tumour types. The PET scans showed that 18 F-FAC accumulated selectively in DCK-positive rather than in DCK-negative tumours (PNAS ). The PET scan offers a preview for how the tumour will react to a specific therapy, explained first author Rachel Laing, a UCLA graduate BSGI proves a valuable option Breast-specific gamma imaging (BSGI) a molecular imaging technology based around a high-resolution gamma camera is proving to be an important tool in the detection and staging of breast cancers. Two recent studies further emphasize the value of this emerging technique, which can visualize lesions independent of tissue density and discover very early-stage cancers. The first of these studies compared the sensitivities of mammography, ultrasonography, MRI and BSGI for detecting invasive lobular carcinoma (ILC). The research team headed up by Rachel Brem, director of breast imaging and intervention at George Washington University Medical Center (Washington, DC) performed a retrospective review of 26 women with biopsy-proven ILC, all of whom had undergone mammography and BSGI (AJR ). medicalphysicsweb review Summer 2009 UCLA/Radu lab gemcitabine sensitive tumour PET probe gemcitabine resistant tumour PET scans: 18 F-FAC accumulates in DCK-positive (on the left side), but not DCK-negative, tumours. researcher in molecular and medical pharmacology. We believe that the tumour cells that absorb the probe will also take up the drug. If the cells do not absorb the probe, it suggests that the tumour might respond better to another medication. The team now plans to examine whether the probe can predict cellular response to other widely used chemotherapy drugs. For the first time, we can watch a chemotherapy drug working inside the living body in real time, explained co-author Caius Radu, assistant professor of molecular and medical pharmacology at UCLA s David Geffen School of Medicine. Radu continued: The beauty of this approach is that it is completely noninvasive and without side effects. We plan to test this method in healthy volunteers within the year, to determine whether we can replicate our current results in humans. If testing in healthy subjects proves safe and effective, the UCLA researchers plan to begin recruiting volunteers for a larger clinical study of the probe in cancer patients. Gamma camera: the Dilon 6800 offers molecular breast imaging. CZT detectors enhance SPECT The mammography and BSGI imaging findings were classified by experienced breast imagers as positive or negative for ILC. Where performed, ultrasound (25 patients) and MRI (12 patients) results were also examined. The researchers concluded that BSGI was the most effective diagnostic imaging technique, with a sensitivity of 93%. Mammography, ultrasound and MRI demonstrated sensitivities of 79, 68 and 83%, respectively. The study is significant because ILC can often be difficult to detect mammographically and is often not palpable at clinical examination, Brem explained. BSGI offers improved detection of this form of breast cancer that impacts approximately 10% of new breast-cancer patients every year. To perform BSGI, the patient is given a radioactive tracing agent. The gamma-ray emission from this radio tracer is then used to form a digital image showing the metabolic activity of the breast tissue. Due to their higher metabolic rate, cancerous cells absorb more tracer than healthy cells and thus appear as hot spots on a BSGI image. BSGI is a physiologic, rather than an anatomic, approach to breast-cancer diagnosis, said Brem. It is likely that the molecular imaging obtained with BSGI is the reason it has the greatest sensitivity for the detection of ILC. In fact, it is known that MRI Wide scope: the team is looking into D-SPECT for areas other than cardiac studies, such as oncology. Most clinical SPECT systems in use today are based on a traditional Anger design of gamma camera: a collimator plus a scintillation detector coupled to a set of photomultiplier tubes. Image acquisition generally involves rotating the detector (or for cardiac studies, dual detectors) around the patient. This scintillatorphotomultiplier combination can be bulky, however, limiting both the speed of rotation and the possible detector location. Replacing the scintillators with a solid-state detector such as cadmium zinc telluride (CZT) enables a more compact design, as well as enhancing the energy resolution. And with the emergence of pixellated CZT detector units now making this option a reality coupled with a global demand for SPECT cardiac perfusion studies Israeli medical device firm Spectrum Dynamics has developed a dedicated CZT-based nuclear cardiology instrument. Now, researchers at University College London (UCL) in the UK have undertaken a detailed study of this so-called D-SPECT system (Phys. Med. Biol ). The aim of this study was to evaluate the performance of a new SPECT system based on novel technology, and compare it to a conventional system based on technology that s been used essentially unchanged for several decades, said lead author Kjell Erlandsson, from UCL s Institute of Nuclear Medicine. While evaluation techniques have been well developed for the traditional system, these are not ideally suited to the new technology. Our challenge was to develop methods that objectively demonstrate the performance of the new system. The D-SPECT scanner comprises nine pixellated CZT blocks, each of which contains pixels with a spacing of 2.46 mm and a total detector area of mm. The compact detectors enable region-centric acquisition, which boosts sensitivity by acquiring data mainly from a preselected region around the heart. The detectors are equipped with wide-acceptance-angle tungsten collimators. The unique features that make pixellated CZT detectors particularly suitable for medical imaging include superior energy resolution, which offers potential advantages in terms of reducing scatter and facilitating multi-radionuclide studies, co-author Brian Hutton told medicalphysicsweb. The detectors are also compact, which allows for a flexible system design and optimization of the data-acquisition geometry. The researchers evaluated a prerelease version of D-SPECT, with comparison scans performed on a conventional dual-headed SPECT (GE Healthcare s Infinia) using standard cardiac acquisition parameters. They first determined the energy resolution of D-SPECT s CZT detectors by obtaining spectra for five different radionuclides. At 140 kev, the D-SPECT resolution was measured as 5.5%, compared with 9.25% for can be limited in the detection of ILC. In addition, the cost of BSGI is significantly less than a breast MRI. Extended view Elsewhere, a research study led by Nathalie Johnson, general surgeon and surgical oncologist at Legacy Good Samaritan Hospital (Portland, OR), demonstrated that BSGI can reveal additional cancers in patients newly diagnosed with breast cancer (Am. J. Surg ). Johnson and her team conducted a retrospective review of 138 breastcancer patients (69 invasive ductal carcinoma, 20 ILC, 32 ductal carcinoma in situ and 17 mixed) in whom BSGI was performed as part of the imaging work-up. BSGI detected additional or more extensive malignancy in the same or opposite breast in 10.9% of these patients. The positive predictive value for BSGI was seen to be 92.9%. According to Johnson, the biggest benefit of BSGI over breast MRI (often used as a mammography the standard gamma camera. The absolute sensitivity of a single CZT detector head (measured using a 99m Tc point-source) was 398 s 1 MBq 1, compared with 72 s 1 MBq 1 for the Infinia s NaI detectors. The team also assessed the D-SPECT s tomographic sensitivity in the heart region, which ranged from 647 to 1107 s 1 MBq 1, depending on the size of the heart being examined and its position in the field of view. The tomographic sensitivity of the Infinia system was 141 s 1 MBq 1. Other measurements included: a scatter fraction of 30% (compared with 34% for the Infinia); an intrinsic resolution equal to the pixel size of 2.46 mm (compared with 3.8 mm); and an average reconstruction resolution, using the standard clinical filter, of 12.5 mm (compared with 13.7 mm for the conventional SPECT). While the D-SPECT exhibited better energy resolution and sensitivity that the conventional system, the researchers believe that its current configuration may not be optimal. They note that the CZT detectors superior energy resolution could be better utilized by using narrower energy windows (currently 20%) to reduce the scatter fraction. What s more, with the reconstructed voxelsize (4.92 mm) currently twice the detector pixel-size, a reduced voxelsize could improve image quality. One specific advantage of the D-SPECT s high energy resolution is the excellent energy separation seen for multiple radionuclides. We have developed a method for simultaneous imaging of 99m Tc and 201 Tl, said Erlandsson. This method, presented at last year s Nuclear Science Symposium and Medical Imaging Conference, has been implemented in the system and is currently undergoing clinical evaluation. adjunct) is its specificity. The sensitivity of BSGI is on par with MRI, but the specificity is higher. In addition, when compared to MRI, BSGI is less expensive and easier to use for patient and physician. She continued: The research is important because it helps clarify the role of BSGI in newly diagnosed breast-cancer patients. We have found that these women can have more extensive disease that is not detected by mammography or ultrasonography. This is especially helpful in patients with dense breast tissue where additional evaluation of the remaining tissue is necessary. In both studies, BSGI was performed with the Dilon 6800 Gamma Camera, a high-resolution, compact field-of-view gamma camera, optimized to perform BSGI. According to its manufacturer Dilon Technologies (Newport News, VA), the camera provides a manageable four to 16 images, compared with up to thousands of images generated with breast MRI. Sign up as a member at medicalphysicsweb.org

11 Invitation for authors Physics in Medicine & Biology The international journal of biomedical physics Fast publication Worldwide visibility High impact Editor-in-Chief: S Webb, Institute of Cancer Research and Royal Marsden NHS Trust, UK If you are working in any of the following areas then we would like to invite your submissions: all areas of radiotherapy physics radiation dosimetry and metrology imaging (e.g. X-ray, MRI, ultrasound, nuclear medicine) image reconstruction and image analysis other radiation medicine applications therapies (non-ionizing radiation) biomedical optics radiation protection radiobiology IMPACT FACTOR * * As listed in ISI s 2008 Science Citation Index Journal citation reports Please submit your research online at For more information visit or us at pmb@iop.org Images: A visual examination of the similarity in plans generated from the nominal non-convex objective function and the relaxed objective function T Halabi, D Croft and T Bortfeld 2006 Phys. Med. Biol keeping you in touch, whenever, wherever... Join the medical physics community and be in with a chance to win an ipod Touch. To enter the prize draw, simply register as a member at medicalphysicsweb.org or complete the registration form on the reverse and return it to booth # 911. The winner will be announced Wednesday morning.

12 12 focus on: diagnostic imaging DIR: validating new algorithms Deformable image registration (DIR) is a key tool for determining a pointto-point correspondence between two clinical images. While several recent publications have described innovative DIR strategies and implementations, the translation of image-processing research into clinical application is confounded by the lack of an established method for validating new algorithms. To determine the efficacy of any DIR algorithm, it s important to compare its performance against that of an appropriate reference standard. Various standards have been proposed, including those based on synthetically deformed images and high-contrast phantoms. But, according to recent research, the best standard for validating registration accuracy in a clinical setting should be one that s derived from actual patient image data (Phys. Med. Biol ). To this end, a US research team has developed a self-contained framework for evaluating DIR accuracy, based on the generation of large samples of expert-determined landmark point pairs from patient scan data. These validation-point sets then act as a reference for characterizing and optimizing DIR algorithms. Our research involves developing and evaluating the accuracy of DIR-based functional lung imaging from 4D CT images, explained Thomas Guerrero, associate professor of radiation oncology at MD Anderson Cancer Center (Houston, TX). It quickly became clear, however, that the derived lung-function images were highly variable, depending on the particular DIR formulation used to perform the pulmonary function calculations. Thus we sought to develop a framework for objective evaluation of DIR spatial accuracy. The landmark point pairs were generated by a thoracic-imaging expert, who manually registered Residual error renderings: a) expert-determined displacement vectors for an example case projected onto a surface rendering of the inhale lung volume; residual error vectors for b) OFM and c) MLS algorithms. large numbers of corresponding pulmonary landmark features from five pairs of treatment-planning thoracic 4D CT images. Using a Matlab-based software interface to facilitate manual selection, the expert registered 6762 landmark point pairs over the five cases. Two secondary readers were employed to provide estimates of observer variance. To demonstrate the practical utility and statistical necessity of the evaluation scheme, the researchers examined two DIR algorithms: a gradient-based optical flow method (OFM) and a landmark-based moving-least-squares (MLS) algorithm. Both algorithms were used to register the five CT image pairs, and the resulting spatial accuracy determined by comparison with the manually registered landmark data set. The mean spatial registration errors for the OFM and MLS algorithms were 6.90 and 2.05 mm, respectively. Graphical representations of these error data provided further detail as to their nature. These algorithms represent two very different strategies for performing DIR of medical image volumes, co-author Richard Castillo told medicalphysicsweb. The OFM formulation is driven by the assumption that corresponding voxel intensities remain constant between image pairs, thus the driving force is image similarity, as opposed to spatial accuracy. The MLS deformation, on the other hand, is guided by input control point pairs, which are intended to anchor the calculation to a subset of spatially accurate grid points. The authors emphasized that the goal of this work was not actually to compare the two DIR algorithms, and note that the poor OFM results could almost certainly be improved via internal parameter optimization. The next stage in this research involves expanding the reference data set further. For the five cases in this study, the researchers plan to propagate the landmarks onto more phases of the 4D CT data sets, providing a means to evaluate the accuracy of full 4D image-registration algorithms. They are also studying a 4D CT image set with corresponding SPECT/CT ventilation and perfusion data. CCTA dose: it s high, but why? Imaging the coronary arteries on 64-slice CT scanners can expose patients to significant levels of ionizing radiation. That statement shouldn t surprise too many medical imaging professionals. What may be more unsettling is the discovery that, while strategies are available for lowering the dose associated with coronary CT angiography (CCTA), these methods are widely under-used (JAMA ). This could do better message comes from an international, multicentre study involving 50 sites (21 university hospitals and 29 community hospitals) and 1965 patients. Participating centres submitted data for all patients undergoing CCTA during one month in the period February to December 2007 (median 31 patients per site). Information given to the study team included the average radiation dose over a specific volume (CTDI vol ) recorded by the CT system, as well as the patient s height, weight and heartbeat, and the reason for their examination. Data analysis was performed by researchers from the German Heart Centre Munich, Germany, together with colleagues from the US Mayo Clinics in Rochester, MN and Jacksonville, FL, and the University of Erlangen, Germany. The researchers calculated the dose-length product (DLP) for each examination to reflect patients radiation exposure during the entire CT scan. The DLP was calculated by multiplying the CTDI vol by the scan length. Effective dose values were estimated by multiplying the DLP by an appropriate conversion factor. Image quality was assessed by an independent reader who checked the visibility of the four main coronary arteries on the CCTA studies. The median DLP for all of the cardiac CT studies turned out to be 885 mgy cm, corresponding to an effective dose of 12 msv or approximately 600 chest X-rays. The typical effective dose for an abdominal/ pelvic CT scan is 10 msv, whilst that for invasive coronary angiography is 5 msv. Most strikingly, the median DLP per site varied enormously - ranging from 331 to 2146 mgy cm illustrating the large variability in CCTA protocols. Linear regression analysis identified four factors that could be modified during CT imaging and which would lower the DLP. These were a shorter scan length, modulation of tube current according to the heart s cycle (25% reduction in DLP), lowering the tube current from 120 to 100 kv (46% reduction), and sequential scanning at one point in the heart s cycle instead of imaging continuously (78% reduction). None of these strategies was used regularly. ECG-controlled tube current modulation was used most often (in 73% of patients). The tube voltage was lowered to 100 kv for 5% of the examinations, and sequential scanning performed in just 6% of cases. In instances where the dose-lowering strategies were used, image quality was not compromised. The study demonstrates that radiation exposure can be reduced substantially by uniformly applying the currently available strategies for dose reduction, but these strategies are used infrequently, the authors wrote. An improved education of physicians and technicians performing cardiac CTA on these dose-saving strategies might be considered to keep the radiation dose as low as reasonably achievable in every patient undergoing cardiac CTA. The degree of variability in CCTA radiation dose between sites has not previously been appreciated, says Andrew Einstein of Columbia University College of Physicians and Surgeons (New York). The results indicate that even centres with considerable experience of CCTA and a high throughput of patients could benefit from quality improvement measures, he noted in an accompanying editorial ( JAMA ). RESEARCH TECHNOLOGY CLINICAL APPLICATIONS A community website from IOP Publishing Your name Title: First name: Last name: address: We need to be able to notify you of your password. Sign-in details Choose a username: Please print clearly, using characters A to Z and 0 to 9 only. I am interested in these specialist areas: Properties of living systems Effects of radiation Monitoring/control/safety Ionizing radiation therapy Non-ionizing radiation therapy Radiation therapy equipment Imaging: general Nuclear medicine X-ray imaging MRI Imaging equipment Biomedical optics Aerospace Radiation measurement Materials and interfaces Biotechniques/instrumentation Nanomedicine Other Please specify Your employment Medical physicist Dosimetrist Clinical scientist Radiation oncologist/therapist Corporate management Engineering management Design/development engineer Research and development Your contact details Institution: Job title: Sales and marketing Consultant Investment/financial Student Radiologist/radiographer Biomedical engineer Quality assurance/quality control Other Please specify Terms and conditions available on request. Thank you for taking the time to register to medicalphysicsweb and entering our competition. Please return the completed form to the IOP Publishing booth # 911. medicalphysicsweb review Summer 2009 Sign up as a member at medicalphysicsweb.org

13 Sharper MR with 7T focuson:diagnosticimaging 13 PTB In contrast: 3 T (left) versus 7 T (right) MR tomography. With equal exam times, the 7 T image shows considerably sharper structures. A 7 tesla MR tomography system at the Max Delbrück Center in Berlin, Germany, is being employed to advance diagnoses and treatments of brain diseases and cancer, as well as open up new possibilities in cardiac research. The machine s ultrahigh field strength greatly enhances image resolution, and enables it to perform high-speed measurements. Project partner PTB, the National German Metrology Institute, is now exploring the technical possibilities of the new device and making it suitable for clinical application. Multiple angles ease screening So far, there are only about 30 magnetic resonance tomographs with such a high magnetic field strength, and most of them are used for brain research, said Bernd Ittermann, head of the PTB s department of medical metrology. The new whole-body 7 T MR machine will have far more widespread use, providing extremely high-resolution images from the interior of the body. The researchers also hope to use the system to investigate molecular processes in the body, to help combat tumours, for example. Early detection of disease means better chance of survival. That s the rationale behind cancer screening programmes. Unfortunately, whenever chest radiography has been trialled as a means of seeking out early-stage lung cancer, mortality rates have not reduced. One reason for the poor performance of chest radiography as a screening tool is its tendency to miss very small pulmonary nodules. Researchers from Duke University (Durham, NC) are now testing a multi-projection X-ray system that promises to do better (IEEE Trans. Nucl. Sci ). Up to 90% of potentially cancerous lung nodules are not seen on standard chest X-rays. Far more small lung nodules are picked up on CT, because the slice-by-slice tomographic technique essentially eliminates the problem of anatomical structures overlapping on 2D views. But the radiation dose associated with a chest CT scan is much higher than that for chest radiography, even if a low-dose protocol is applied. Implementation of a lung cancer screening programme based on CT consequently remains controversial. From a public health standpoint, the higher associated health risks might not be justifiable for a widescale CT lung cancer screening program. Thus if lung cancer screening is to be realized, there is a need for a cost-effective and dose-effective alternative, say Duke University s Amarpreet Chawla and colleagues. Multi-projection radiography offers a way to improve the detection of pulmonary nodules without raising the overall dose above that of a standard chest X-ray. The idea is to acquire multiple digital radiographic images in quick succession from slightly different angles over a limited angular range. Information gathered from each projection is then combined. This reduces the confounding effects of anatomical overlap and increases the chance of pathology being seen. The clinical value of such a system, designed and built at the Duke University radiology department, is now being tested on a cohort of normal volunteers and patients with confirmed lung nodules. The core of the system hardware is a standard X-ray unit with an upright gantry. This has been modified so that the X-ray tube translates along the horizontal and vertical axes in an iso-centric motion. Multiple images can be acquired from different angles in rapid succession in both step-andshoot and continuous modes. For each study subject, three projections will be acquired along the horizontal axis at +3, 0 and 3. The dose for each acquisition, which takes 10 s to complete, will be onethird that of a typical chest X-ray. Results from the first 80 individuals to be imaged with the multiprojection system have now been released. Experienced radiologists rated the diagnostic image quality as excellent, Chawla and colleagues report. No motion artefacts due to the system were observed either. Preliminary clinical trials indicate that the system, used as a diagnostic tool, could potentially be used to improve detection of lung nodules, they concluded. PENTA-GUIDE ENSURE DAILY ON-BOARD IMAGING TARGETING ACCURACY Test linac-mounted volume imaging systems for: 3D cone beam registration kv and MV system coincidence kv and MV projection images Laser and light field coincidence Remote table adjustments ACCURACY IS CRITICAL Visit us at AAPM booth #608 Modus To find out more about advertising opportunities on medicalphysicsweb and in the medicalphysicsweb review, simply contact our sales representative Paul Rucci or see us today at booth # rucci@ioppubusa.com, office tel: +1 (215) ext. 268, cell number: Sign up as a member at medicalphysicsweb.org medicalphysicswebreviewsummer 2009

14 14 focuson:nanomedicine Nanoparticles: image, target, treat Nanoparticles in their myriad of shapes and forms are touted as ideal for a broad range of cancermanagement applications. In a hotbed of research activity, nanoscalematerials are being developed to image the distribution of tumour cells in the body, target and attach to cancerous cells, destroy unwanted cells via ablation or delivering drugs, or according to the latest claims all of the above. Here, we take a look at some of the nanotechnology developments reported recently. Royal Philips Electronics of the Netherlands has created the first in vivo 3D images using magnetic particle imaging (MPI) a technique based on visualizing iron-oxide nanoparticles injected into the bloodstream. The team used MPI scanning to create real-time images of arterial blood f low and volumetric heart motion in a beating mouse heart. By combining high spatial resolution with short image acquisition times, MPI can capture dynamic concentration changes as the nanoparticles are swept along by the blood stream (Phys. Med. Biol. 54 L1). By adding important functional information to the anatomical data Luo Gu obtained from existing modalities, such as CT and MR, Philips MPI technology has the potential to significantly help in the diagnosis and treatment planning of major diseases such as atherosclerosis and congenital heart defects, said Henk van Houten, senior vice-president of Philips Research. Many nanomaterials tested in research labs are too toxic for human use. Now, a team of US scientists has developed porous silicon nanoparticles that glow brightly, last long enough to slowly release cancer drugs, and then break down into harmless by-products. When the nanoparticles were tested in mice, the tumours glowed for several Safety first: silicon nanoparticles glow red under ultraviolet light. hours and then faded. Close examination of vulnerable organs like the liver, spleen and kidney revealed no lasting changes in mice treated with the nanoparticles (Nature Materials doi: /nmat2398). The biodegradable nanoparticles can also be used to help deliver cancer drugs. The goal is to use the nanoparticles to chaperone the drug directly to the tumour, said study leader Michael Sailor of the University of California, San Diego (La Jolla, CA). At about 100 nm, these particles are bigger than many others designed to deliver drugs, a factor that s claimed to improve both their effectiveness and safety. A US research team is using hollow gold nanospheres to enhance the cell-killing effects of photothermal ablation. The ultimate aim is to develop a minimally invasive treatment for malignant melanoma. The researchers equipped the nanospheres with a protein fragment that targets melanoma cells while avoiding healthy skin cells. When exposed to near-infrared light, the nanospheres heat up and destroy the cancer cells. In studies in mice, the hollow nanospheres did eight times more damage to skin tumours than the same particles without the targeting peptides. They re also claimed to be 50 times more effective than solid gold nanoparticles for photothermal therapy. With sizes ranging from 30 to 50 nm, the nanoshells are much smaller than other nanoparticles previously designed for photoablation therapy. The next stage will be to test the nanospheres in humans. Though with the need for extensive preclinical toxicity studies, there s some way to go yet before the technique could find its way into clinical practice. The researchers, from the University of California Santa Cruz, the University of California Berkeley, and the MD Anderson Cancer Center in Houston, TX, presented this work at the 237th ACS National Meeting in Salt Lake City, UT. Researchers at Purdue University (West Lafayette, IN) have come up with a way to deliver drugs directly to cancer cells using gold nanorods combined with magnetic iron-oxide particles. The magnetic particles can be traced via MRI, while the nanorods are luminescent and can be traced through optical microscopy. Purdue Agricultural Communication photo/tom Campbell While MRI is less precise than optical luminescence in tracking the probes, it can track them deeper in tissue (Angew. Chem. Int. Ed ). On target: Irudayaraj displays the particles magnetic attraction. The probes contain the antibody Herceptin, which locates and attaches to protein markers on the surface of cancer cells. The team has tested the probes in cultured cancer cells and now plans experiments in mice models to determine the dose and stability of the probes. When the cancer cell expresses a protein marker that is complementary to Herceptin, then it binds to that marker, said Joseph Irudayaraj, a Purdue associate professor. The probe could carry drugs to target and treat, as well as reveal cancer cells. We are at AAPM booth number 676 and ASTRO booth number 2907 Introducing DVS The only implantable dosimeter for dose verification and target localization. See more. Learn more. Visit us at AAPM and ASTRO. medicalphysicswebreviewsummer 2009 Sign up as a member at medicalphysicsweb.org

15 The Unfors Xi system can simultaneously measure kvp, dose, dose rate, HVL, pulse, pulse rate, dose/pulse, time and waveforms with the R/F & MAM detector. The Unfors Xi, now in its Platinum Edition, is a complete system for diagnostic x-ray multi-parameter measurements on all modalities. This includes Radiography, Fluoroscopy, Mammography, Dental and CT. Added options include separate detectors for luminance and illuminance measurements and radiation scatter or leakage measurements, both of which can be quickly attached to the Unfors Xi. For mammography applications, the Unfors Xi provides a unique feature to measure HVL in only one exposure without the need to use filters. The M-Pro mammography option offers calibrations for the new W/Ag and W/Rh beam qualities. For reporting purposes, the new Unfors QA View software is now available. Flexible reporting software One meter all applications Truly easy to use All designed to increase your accuracy and productivity! The Unfors Concept Accurate result 10s to learn Pocket sized Unfors Instruments, Inc. 48 Anderson Avenue, Suite 1 New Milford, CT 06776, USA Phone: (866) 4UNFORS +1 (860) Fax: +1 (860) info@unfors.com

16