| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Review article |
Departments of 1 Radiotherapy Physics and 2 Clinical Oncology, The Ipswich Hospital NHS Trust, Heath Road, Ipswich, Suffolk IP4 5PD, UK
 |
Abstract |
|---|
 Top Abstract Clinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
|
Clinical role of IMRT |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
There is a misconception among some patients that IMRT will be superior to 3D CRT for their tumour. This may be true in some cases, but certainly not all. The tour de force with IMRT is in treating the concave target volume. Data have been published on potential advantages of IMRT treatments over conventional treatments in planning studies for selected tumour sites [6–13]. The main rationale in these sites, predominantly head and neck, is better conformality of the target volume whilst sparing radio-sensitive normal tissues. IMRT also has a role as a missing tissue compensator as has been shown for bladder treatments [14] and in treatments of the breast [15, 16] by compensating for the irregular shape of the patient.
At the Ipswich Hospital, implementation of IMRT commenced in June 2001 and a number of patients have now been treated with inverse planned, multiple-field IMRT and more simple forward planned electronic compensation techniques. This paper describes our experiences as we established IMRT as a routine treatment within the department.
|
Implementation of IMRT in a district general hospital |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
During 1999 it was acknowledged that replacement equipment was required for the existing two treatment machines. By this time 3D CRT was well established within the department with some patients with head and neck, lung and gastrointestinal tumours benefiting along with those with prostate cancer. To minimize the disruption to the department, funding was made available for two new high-energy bunkers to be built to house the new machines. It was felt that in order to further improve and develop our radiotherapy service the new equipment should be capable of delivering intensity-modulated treatments. In January 2001 two Varian 2300EX linear accelerators (Varian Medical Systems, Palo Alto, CA) were installed. The machines are beam-matched to enable transfer of patients without the need for re-planning. Each machine has 6 MV and 18 MV photons, a range of electron energies, 120 leaf MLC with dynamic capabilities for IMRT treatments and amorphous silicon electronic portal imaging (aSi EPI). Also included in the new equipment purchase was a Varian CadplanTM 6.2.7 3D TPS with HeliosTM inverse planning module linked to a Varian SomaVisionTM contouring platform. The TPS was networked to the treatment machines via the Varian VarisVisionTM record and verify system, enabling complete electronic transfer of data. Selecting equipment from the same manufacturer minimized connectivity issues and ultimately facilitated a relatively problem-free commissioning period.
The inverse planning algorithm [19] incorporates a conjugate gradient method to create optimum beam fluences from dose volume constraints and their relative priorities as defined by the user. These optimal beam fluences are then converted to actual deliverable beam fluences using a leaf motion calculator (LMC) [20] also incorporated in the software. The user is able to select either a sliding window dynamic MLC delivery, where the MLC leaf pairs are swept from left to right over the treatment field with varying aperture whilst the beam is on, or a step and shoot delivery, where a series of static fields or segments are delivered to build up the required fluence for each field. The user can define the number of intensity levels for the step and shoot deliveries. When calculating leaf motion files the LMC incorporates maximum leaf speed, machine dose rate and takes into account a minimum leaf gap to ensure there are no collisions between opposing leaves. Leaf transmission and leakage through the rounded leaf ends, known as dosimetric leaf separation [21] are also utilized by the LMC.
Commissioning
Acceptance testing of the two linear accelerators was completed in February 2001 and the commissioning programme commenced, which from the outset included verification of the IMRT planning and delivery systems. A series of tests were undertaken to ensure the integrity of the system as a whole. They were designed to establish confidence in the equipment and form a basis for routine quality assurance procedures.
Testing dynamic MLC
The initial tests performed were used to ensure the accuracy and calibration of leaf positioning and carriage movement, to ensure the dynamic MLC accurately controlled the beam and to measure the dosimetric leaf parameters required by the LMC. Leaf transmission and interleaf leakage, and dosimetric leaf separation were measured following methods described by LoSasso et al [21] and the results compared favourably with values recommended by the manufacturer.
Planning and delivery studies
A hypothetical prostate cancer treatment was created on the TPS. The inverse planning software was used to optimize the fluence of each of five fields to satisfy pre-defined criteria for target dose and dose to organs at risk. Deliverable fluences incorporating physical leaf parameters were calculated, two sets per field, one for sliding window delivery and a second for step and shoot. The fluences were exported to a water phantom within the TPS, ensuring the position of the isocentre and jaw sizes remained the same for each field. Line profiles for each field at isocentre level and isodose plots were calculated by the TPS. The plans were then exported to the linear accelerator for delivery. Fluence measurements and line dose measurements at a reference depth were made using film (Kodak X-OmatVTM, Eastman Kodak Company, Rochester, USA) and an LA48TM Linear Chamber Array (PTW-Freiburg, Freiburg, Germany) for comparison with isodose plots and line profiles calculated by the TPS. Absolute dose measurements using a calibrated PinPointTM 0.015 cm3 ionization chamber (PTW-Freiburg, Freiburg, Germany) were also made in water for comparison with calculated doses. There was good agreement between measured and calculated fluence maps for both the sliding window and step and shoot deliveries, with maximum discrepancies of 2 mm in high dose gradient regions and 3% elsewhere. There was also good agreement between measured and calculated line doses (Figure 1
), with the sliding window delivery showing slightly better agreement (approximately 1% maximum discrepancy for the sliding window and 2% for the step and shoot delivery). This may be due to having restricted the intensity levels of the step and shoot delivery to five. The absolute dose measurements showed measured doses to be up to 2% lower than calculated for the sliding window delivery and up to 4% lower for the step and shoot delivery.
|
External audit
At the time of commissioning we were fortunate enough to have the Standardization of Breast Radiotherapy (START) Trial quality assurance (QA) Team in the department. Consequently the START trial anthropomorphic breast phantom [22] was used to measure absorbed dose at multiple points within the breast volume and verify the dose delivered by a two field tangential breast plan with electronic compensation delivered by sliding window MLC. The results were very encouraging, with the measured data consistently 2% to 2.5% lower than calculated, tying in well with our own phantom measurements.
Training
All staff require training with any new technology and we felt this preparation needed to commence well in advance of the delivery of the new hardware, the constant throughput of patients contributing to a lost opportunity if this approach was not taken. The implementation of the IMRT service was also an opportunity to examine and define more explicitly the roles of the various team players. Historically radiotherapy has always been multiprofessional in its delivery though the exact roles of staff in individual departments will vary, often dictated by the skills of those in post at the time.
The introduction of IMRT was to be a steep learning curve for all staff groups and required additional training and professional development. We felt such training would be best delivered by centres with the clinical experience and expertise in IMRT using equipment from the same manufacturer. Such courses have been valuable in being both machine-specific and tailored to both physicists and clinicians. In addition, key personnel have profited from the annual European Society for Therapeutic Radiology and Oncology (ESTRO) course on "IMRT and Other Conformal Techniques in Practice". At the time of our initial training, courses were only available in the USA and Europe and it is encouraging to see that a dedicated IMRT course is now being run in the UK at the Royal Marsden Hospital.
Staff training and development is an on-going process and regular department meetings with involvement of all professional groups in presentations are pivotal in maintaining the momentum.
Patient selection and dual planning
Our aim when introducing IMRT to our department was to improve on our established 3D CRT technique by delivering a more conformal dose to irregularly shaped target volumes and reducing the dose to organs at risk, potentially resulting in better tumour control and lower complication rates. It seemed that any tumour site should be considered for IMRT where there was the likelihood of normal tissue tolerance being exceeded using our standard technique and where a radical treatment was otherwise indicated. Therefore our underlying principle has been of perceived superiority of the IMRT plan over 3D CRT. Consequently patients were initially dual-planned.
|
QA for IMRT |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
lists the series of dynamic MLC checks that are performed periodically by radiotherapy technicians. These checks form part of the linear accelerator QA procedures.
|
Cumulative dose measurements
On the TPS the patient plan is exported to a CT scan of a Perspex block phantom and the dose recalculated maintaining isocentre position, fluence and MUs for each field. A suitable reference point is selected in an area of minimal dose gradient, and its point dose calculated. The patient plan is exported to the record and verify system for treatment and delivered to the Perspex block phantom in its entirety with the linear accelerator running in clinical mode. The cumulative dose to the reference point is measured with a calibrated 0.015 cm3 ionization chamber. The measured dose is then compared with the TPS calculated dose.
MU check
On the TPS each planned field with associated fluence is exported individually to a water phantom with gantry angle 0° and 95 cm surface–skin distance (SSD). The dose at 5 cm deep at isocentre is calculated. With the gantry angles changed to 0° for each field on the record and verify system, the fields are delivered in turn to a solid water phantom with its surface at 95 cm SSD. A 0.015 cm3 ionization chamber positioned at isocentre in the phantom measures the dose delivered by each field for comparison with the TPS calculated doses. Measured doses are then corrected back to the actual treatment depth for each field by the tissue phantom ratio (TPR) and combined to give the total measured dose to the isocentre. This is compared with the planned dose as given by the TPS and acts as an independent MU check.
Individual fluence measurements
The 0.015 cm3 ionization chamber is replaced by Kodak X-OmatV film at 5 cm deep, 95 cm SSD in the solid water phantom. Each field is delivered to film and a series of calibration films are irradiated under the same conditions to known doses. The processed films are digitized with a Vidar VXR16TM film scanner (Vidar Systems Corporation, Herndon, USA) and analysed with MEPHYSTOTM 7.3 therapy beam data acquisition and analysis software (PTW-Freiburg, Freiburg, Germany). The calibration films provide a plot of optical density against dose which is then used to correct the treatment field films and provide absolute isodose plots for the 5 cm reference depth. The plots are compared with isodoses calculated by the TPS. No quantitative analysis of these isodoses is presently undertaken however and the two sets of isodoses are overlaid on a light-box to give an indication of any differences.
Cumulative fluence measurements
In order to obtain an indication of the delivered fluence in the transverse plane, a Kodak Extended Dose Range (EDR2TM) film (Eastman Kodak Company, Rochester, USA) is placed upright between slices of Perspex on the treatment couch at isocentre level. The planned sequence including appropriate gantry angles is delivered to the film, which is then processed and digitized. Relative isodoses are plotted for comparison with isodoses calculated by the TPS.
Electronic compensator QA
On the TPS each field is exported separately to the water phantom where dose is calculated at the isocentre 5 cm deep. Isodoses are also calculated and plotted at this depth. Individual leaf motion files are exported from the TPS to the linear accelerator and delivered to a solid water phantom with its surface at 95 cm SSD. MU and jaw positions are maintained for the delivered sequences. The dose at isocentre is measured with a 0.015 cm3 ionization chamber for comparison with calculated dose and corrected back to the treatment depth by the TPR to act as a MU check. Kodak EDR2 film is used to measure the absolute beam fluence for comparison with the calculated fluence.
On-treatment verification
Electronic portal imaging
The aSi EPI systems have led to the development of radiographer led protocols for on-treatment image acquisition, analysis and correction of patient position. Each CT planned patient has a separate "isocentre check plan" created on the TPS consisting of orthogonal 10 cm x 10 cm fields, usually with gantry angle 0° and 90° and with the same isocentre as the treatment plan. Isocentre check images are acquired during the first three fractions and at weekly intervals thereafter and can show discrepancy in isocentre position in anterior/posterior, superior/inferior and medial/lateral directions. A maximum of 3 MU per image are required which is not incorporated in the planned dose. Quantitative image analysis is performed by trained radiographers using the Varian PortalVisionTM software, comparing portal images with DRRs from the TPS. Field edge plots (Figure 2) created for the first three fractions show clearly any systematic error in isocentre position. Tolerances of 3 mm are specified for 3D CRT and IMRT treatments and systematic errors are corrected for by moving the patient as necessary.
|
is a simulator film showing the location of a TLD used to assess the dose within the target volume of a patient with an olfactory neuroblastoma. Entrance doses on the central axis for the electronic compensator breast fields are measured using calibrated p-type silicon diodes (EDP-15, Scanditronix/Wellhofer, Schwarzenbruck, Germany). Reference doses are taken from the TPS and compared with diode readings corrected for temperature and angular dependence.
|
Clinical skill required for IMRT
All staff groups need to be familiar with the principles behind IMRT, which to some extent can be gained with experience of 3D CRT and an understanding of how IMRT can be considered as the evolution of such techniques.
The radiotherapist
There is no obvious innate skill incumbent upon the treating clinician but a well-practised ability in CT planning is fundamental to both a 3D CRT and IMRT approach in radiotherapy. True CT planning means delineating the target volume on individual slices. The implementation of CT planning in a department depends very much upon the skills level. We adopted a step-wise approach in the management of head and neck cancer as has been described previously [27]. Initially the traditional bony landmarks served as the "quality assurance" check in outlining on an axial CT scan. Confidence in the technique however has placed less reliance on this, recognising that disease may actually be missed on occasions if this approach is followed too rigidly. We have found it is essential that CT planning should be used for all phases in a multiphase technique. It can be possible to delineate the phase one volume as a standard field only to find that the boost volume that was carefully defined on CT actually lay beyond the original field limits! The complexity of target volume delineation may require collaboration across individuals' areas of subspeciality expertise in addition to work with radiological colleagues already described. For example, in our institution, head and neck lymphoma patients are planned and treated by the head and neck oncologist but ultimately managed by the clinical oncologist responsible for haemato-oncology. The clinician must also be able to review both the resultant dose distributions and pre-treatment QA of IMRT plans with physics staff with an understanding of the implications of what is being shown. This should not be seen as a "signing-off" exercise but more as an example of a multidisciplinary approach to effective radiotherapy treatment.
The radiographer
With the introduction of 3D CRT the radiographer's role was extended in both treatment planning and delivery. The RCR recommends interdisciplinary working following appropriate additional training [28] and in our own institution, breast outlining and most normal tissue outlining is performed by trained planning radiographers with the proviso that critical normal tissues are independently verified by the clinician. Schools of radiography include diagnostic radiology in their courses, which is invaluable as CT planning becomes more common. The tolerances required in setting up patients for IMRT may be alien to some radiographers without a good grounding in 3D CRT, though in many respects the need for the utmost reproducibility is nothing more than the pursuit of excellence for all radiotherapy patients. Our long established 3D conformal service in prostate cancer engendered this culture amongst qualified and student radiographers. A further necessary skill for treatment radiographers is review and analysis of EPIs. Again this skill has been gained through 3D CRT and directly applied to IMRT treatments.
The physicist
As with the radiotherapist and radiographer, a good grounding in all aspects of 3D CRT is essential for the physicist as is an appreciation of the aims and limitations of IMRT. In our centre the physicist's role incorporates inverse planning and pre-treatment QA as well as advising other staff groups in all aspects of IMRT planning and delivery. A thorough understanding of inverse planning and optimization is required and as well as familiarity with the TPS software in order to develop the most appropriate way to drive the optimization. This is system dependent and can only be gained with experience. The physicist needs to understand the requirements for both machine specific and patient specific QA, to be able to interpret results and communicate these results effectively with clinical and technical staff.
It goes without saying that teamwork is crucial to the safe implementation of IMRT and an understanding of everyone's role should be explicit at the outset.
|
Clinical experience with IMRT |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
Some lung tumours have also been treated with IMRT. Those selected have been paraspinal volumes where, were it not for spinal cord overdose, radical treatment would have been indicated. The use of 3D CRT and IMRT for such volumes has been previously reported by Pirzkall et al [33]. As in our head and neck cases, we have not made any initial changes to our fractionation, the fundamental difference being that patients who previously would have been treated palliatively can now be given a radical dose.
Our breast experience has evolved along similar lines. IMRT has been used as an electronic compensation technique only, aiming to improve dose homogeneity within the breast. Beam arrangements and dose fractionation remain unchanged for these treatments, the only difference being the introduction of routine CT planning.
Treatment planning
For prospective IMRT patients the journey through pre-treatment is the same as for any CT planned radical treatment. Patient positioning and immobilization are unchanged with the introduction of IMRT as we deemed our existing methods developed for 3D CRT treatments to be suitable. The only significant change has been for breast patients where the introduction of CT planning prompted a change in treatment position. Patients are scanned and treated on an angled board with both arms abducted to level the sternum.
Each breast patient is dual planned for both a conventional enhanced dynamic wedged (EDW) treatment and electronic compensation. Optimization of the EDW plan is primarily based on the resulting dose distribution in the transverse mid-plane following the START nationally agreed planning protocol [34]. For the electronically compensated plans, the TPS calculates appropriate beam fluences to achieve an even dose distribution in the mid-plane sagittal contour. Leaf motion files are then calculated to deliver these fluences by the sliding window technique. All plans are normalized to the START reference point and the optimal plan selected for treatment. Plan selection is based on the 3D dose distribution within the whole breast volume and quantified by the maximum dose within the treated volume, the volume of tissue receiving more than 105% of the dose, and the coverage of the planning target volume (PTV) [35].
For the inverse planned head and neck and paraspinal lung treatments five or seven equally spaced coplanar fields are applied with an anterior or posterior field always included. The isocentre for the plan is normally the geometric centre of the PTV and the collimator rotation is maintained at 0°. Field orientation may be reconsidered after optimization depending on the position of hot spots within the calculated plan although no computerized beam direction optimization is undertaken.
The leaf motion pattern for each beam is generated from the optimal beam fluence for a sliding window delivery and actual fluence is calculated for each beam along with the 3D dose distribution. Resulting plans are evaluated by the clinician and physicist planning the treatment. Dose–volume histograms (DVH) and graphical representations of the dose distribution are invaluable tools for assessment and comparison of IMRT plans with 3D CRT plans for the same patient.
Treatment delivery
Once an IMRT plan has been selected for treatment, the QA undertaken and simulator verification of patient position has taken place, the plan is sent to treatment. We have found it useful to carry out a "dummy-run" to check the feasibility of the treatment, to detect any unexpected problems and ensure auto-field sequencing on the linear accelerator can be used. Each plan takes less than 1 min to download to the treatment machine and actual delivery times are 10–15 min for a 7 field treatment and 3–5 min for electronically compensated breast treatments.
|
Results of IMRT |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
shows the resulting mean discrepancies for all plans between measured and calculated doses and the maximum discrepancy observed between measured and calculated fluence. These results are consistent with those obtained during commissioning measurements. A series of four consecutive plans with measured dose lower than expected contributed to the relatively large standard deviation in the cumulative dose measurements. In conjunction with dynamic MLC QA undertaken at the time, these unexpected results led to re-calibration of the MLC and pre-treatment verification of subsequent plans showed improved discrepancies between measured and calculated cumulative dose of <3%.
|
Electronic compensator QA
At the time of writing a total of 202 electronically compensated breast plans have undergone pre-treatment verification and 99% of individual field dose measurements were within ±2% of calculated (all within ±3%). When converted to dose at the treatment depth using the TPR, the discrepancy between measured and TPS calculated dose at the isocentre was <±2%. Measured fluences showed no significant difference from calculated. Pre-treatment verification has taken approximately 15 min treatment machine time and a further 15 min for preparation and analysis per patient. The results have given us full confidence in the LMC and dynamic MLC delivery systems for electronically compensated treatments.
Head and neck treatments
Case selection
12 cases have been treated with IMRT to date (Table 3). The primary reason for selecting IMRT for these patients was to achieve adequate coverage of the CTV whilst sparing normal tissues to at least within conventional tolerance limits [36, 37]. Where applicable, node levels were outlined according to guidelines available at the time [38–40] and involved nodes were outlined to aid in target volume delineation (Figure 4).
|
|
. In one case, the treatment was undertaken in order to achieve maximal local control despite the presence of metastatic disease which is currently receiving radioisotope therapy.
|
Tumours within the paranasal sinus/nasal cavity have included olfactory neuroblastoma managed initially by surgery and chemotherapy, and adenoid cystic carcinoma with extensive bony and perineural invasion. In such cases, IMRT planning allowed delivery of the intended dose to the target volume with maintenance of neuronal tissues within conventional tolerance limits. It was not always possible to reduce doses below such limits and we attribute this to the close proximity of the CTV and PTV to the relevant organs at risk and the inevitable dose gradient. A further reduction in normal tissue dose might have been possible at the expense of under-dosing the CTV which was not our intention.
Parotid-sparing radiotherapy was given to two patients who either explicitly sought it or expressed a preference on informed consent. It was acknowledged at the time that the technique of parotid-sparing was strongly advocated by some [41] though the subjective benefit may not be quite as one would hope [42]. In order to ensure that such treatment would be appropriate for the patient, baseline salivary function was sought by scintigraphy [43] which would also serve as an objective measure post-treatment if required. The mean dose to the parotid gland intended to be spared was kept below 26 Gy in accordance with Eisbruch et al [44].
We have utilized IMRT for one case of non-Hodgkin's lymphoma where there was involvement of the nasopharynx. Despite the low doses delivered, we encountered considerable difficulties in encompassing the CTV which by necessity included Waldeyer's ring, the nasal cavity and the tumour's extension into the infratemporal fossa.
Acute reactions
All patients were monitored for unusual acute reactions. This was especially important given the multiple beam arrangements often exiting through structures not normally irradiated even to moderately low doses with conventional techniques. In the case of thyroid cancer, there was an unusually brisk skin reaction. Though it may have been due to intrinsic radiosensitivity, we attribute this to the inclusion of the skin within the grown PTV. This has recently been reported elsewhere [45] and as a consequence of this, the PTV is now modified routinely to ensure that it is no closer than 3 mm to the skin surface unless considered at risk. Significant nausea was experienced by two patients during treatment necessitating subcutaneous antiemetics and settled soon after completing the radiotherapy. Both patients were inpatients with brisk pharyngeal reactions and the actual cause of the nausea is therefore unclear. Except in one patient where the orbit was irradiated to full dose due to tumour involvement resulting in the expected acute reaction, no patient experienced anything more than mild conjunctival irritation when the beam arrangement included the orbit. In all such cases, the lacrimal gland was treated to less than 40 Gy to minimize late effect [46] and similarly the lens to less than 10 Gy. In vivo dosimetry supported these intended dose limits. Alopecia occurred as expected according to the beam arrangements.
Tumour control
One patient has developed an in-field recurrence and died, one has relapsed out-field. No other patients have recurred to date though one case of medullary thyroid cancer is currently under evaluation for an edge recurrence. Follow up is short however for these locally advanced tumours.
Late effects
No patient has experienced late toxicity thus far. Where the intention of IMRT was to reduce xerostomia, there is objective evidence of sparing of the relevant salivary gland as shown in Figure 6.
|
Inverse planned IMRT was the treatment of choice due to the proximity of the spinal cord to the target volumes and the plan optimization was driven in order to keep the maximum cord dose to <46 Gy over the whole treatment. Strict dose volume constraints following those given in Hernando et al [47] were also placed on the lung volumes with both lungs being viewed as a single structure. Achieving a homogeneous dose within the target volume was not considered as important as keeping the minimum target dose above 95% and ensuring the position of any hot-spots (>105%) were not in close proximity to the spinal cord or untreated lung. Figure 7 shows the dose distribution achieved in the transverse plane for a paraspinal volume.
|
).
|
Palliative treatments
One patient with metastatic disease of the pleura following primary breast cancer has been treated palliatively with IMRT. The aim was to deliver a dose of 50 Gy to the pleura whilst sparing some function in the affected lung (Figure 8) and improve the patient's quality of life by reducing breathlessness and other respiratory symptoms. Although the patient found the treatment difficult to tolerate at times, follow-up has shown good results and we feel, justified our choice of plan.
|
|
The future |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
Without question we see the increased utility of IMRT within the head and neck territory. In addition to facilitating improved coverage of the CTV and sparing of organs at risk, the use of IMRT has avoided field matching with electrons that is necessary in conventional multiphase treatments. Some patients have found the treatment times onerous, but the cumulative times are actually less than with conventional treatments. IMRT can facilitate altered total doses within carefully defined volumes considered at variable risk. This approach has already been published [50] and more mature results are emerging [51]. The "synchronous boost" technique facilitates accelerated fractionation as a means of improving outcome and reduces pre-treatment verification as only one treatment phase is required. We have been cautious in utilizing this technique to date but expect to use it more readily in future as we grow more comfortable with the radiobiological aspects of altered dose per fraction.
We feel that using IMRT to enable treatment of paraspinal lung tumours with radical doses is safe in our limited experience. We predict the use of IMRT for such "fixed" lung volumes could possibly lead to an accelerated dose per fraction and the possibility of combining the treatment with concurrent chemotherapy. However, this would only be introduced here within the confines of clinical trials.
Electronic compensation of tangential breast fields has improved the dose homogeneity within the breast for around two-thirds of our CT planned breast patients. Such treatments take no longer to plan or deliver than conventional wedged although it is difficult to predict which patients will benefit from the technique. Consequently dual-planning such treatments will continue with the exception of chest wall treatments where the electronic compensation offers no improvement.
The one major site where we have not yet utilized IMRT is within the pelvis. At the outset of our program there was emerging data from the USA of the superiority of IMRT in prostate treatments when escalating doses towards 80 Gy and above [52]. At the time our standard prostate dose was 64 Gy and our 3 field 3D CRT technique was adequate at keeping rectal doses below our specified tolerance. Dose escalation for prostate cancer is on a firmer footing now [53] and since our participation in the MRC RT01 prostate trial, 74 Gy has become our routine prescribed dose. However, due to strict adherence to our DVH criteria, a significant proportion of patients cannot benefit from this higher dose using our conventional 3D CRT plan. In order to achieve the improved outcome expected with dose escalation we intend to explore the clinical utility of IMRT for prostate cancer. Indeed, Corletto et al [54] have recently published work that supports the philosophy that IMRT is complementary to 3D CRT and of most value in prostate cancer where the target volume is concave. We also feel IMRT has potential benefits for gynaecological and colorectal treatments and these are likely to be incorporated in our IMRT programme.
We had not expected palliative treatments to benefit from IMRT but one case has now been undertaken. Though we are mindful of our underlying principle that IMRT will compliment our current techniques there is data that IMRT will be of value in diverse sites, e.g. mesothelioma [55, 56] and vertebral body metastases [57]. Notwithstanding the ever-present resource issue, we will continue to use IMRT in the palliative setting where there is going to be a real clinical benefit.
To date we have we have undertaken no optimization of beam number or beam orientation for the inverse-planned treatments, predominantly selecting seven equally spaced fields which have resulted in satisfactory dose distributions. However, studies have shown the importance of beam number and orientation for some sites [58, 59] and this will be explored as our IMRT programme continues.
We anticipate that the impact of pre-treatment verification on physicist and treatment machine time and resources will diminish as confidence in the planning and delivery systems grows. In recent months commercially available software for an independent MU check (RadCalcTM, LifeLine Software Inc., TX, USA) has been commissioned and used in parallel with the pre-treatment verification for both inverse-planned IMRT and electronic compensation. Consequently we have taken the step of ceasing pre-treatment verification for electronic compensation treatments, although a rigorous check of the plan and use of the RadCalc software by a physicist will continue. For the inverse-planned IMRT treatments, pre-treatment verification will continue in its current form. We are currently commissioning the MapCHECKTM two-dimensional diode array (Sun Nuclear Corporation, FL, USA) with a view to replacing film with the device for fluence checks. It is hoped that the use of MapCHECK in conjunction with the independent MU check software will enable streamlining of the pre-treatment verification.
The impact of expanding the IMRT service on the clinician remains unclear as the need to make more use of different imaging modalities and the inevitable attention to detail here could conceivably increase the workload. However, confidence in the technique and familiarity of the current and expected guidelines on target volume delineation [60] has made some aspects of the task easier.
The implementation of IMRT at the Ipswich Hospital has been possible with the appropriate equipment and the enthusiasm and motivation of a multidisciplinary team. Expansion of the programme will continue within the bounds of available resources and in many cases within the confines of multicentre clinical trials.
|
Acknowledgments |
|---|
Received for publication June 24, 2003. Revision received August 15, 2003. Accepted for publication September 2, 2003.
|
References |
|---|
|
TopAbstractClinical role of IMRTImplementation of IMRT in...QA for IMRTClinical experience with IMRTResults of IMRTThe futureReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C H CLARK, E A MILES, M T G. URBANO, S A BHIDE, A M BIDMEAD, K J HARRINGTON, C M NUTTING, and on behalf of the UK PARSPORT Trial Management Grou Pre-trial quality assurance processes for an intensity-modulated radiation therapy (IMRT) trial: PARSPORT, a UK multicentre Phase III trial comparing conventional radiotherapy and parotid-sparing IMRT for locally advanced head and neck cancer Br. J. Radiol., July 1, 2009; 82(979): 585 - 594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P P Dendy The President's Conference 2005: "Technology in Imaging and Radiotherapy - towards improved workflow and productivity" Br. J. Radiol., January 1, 2006; 79(937): 1 - 4. [Full Text] [PDF]
|
|
|
|
|
|
S Webb Intensity-modulated radiation therapy (IMRT): a clinical reality for cancer treatment, "any fool can understand this": The 2004 Silvanus Thompson Memorial Lecture Br. J. Radiol., October 1, 2005; 78(Special_Issue_2): S64 - S72. [Full Text] [PDF]
|
|
|
|
|
|
BJR Review of the Year - 2004 Br. J. Radiol., March 1, 2005; 78(927): 181 - 185. [Full Text] [PDF]
|
|
|
|
|
|
A W Beavis Is tomotherapy the future of IMRT? Br. J. Radiol., April 1, 2004; 77(916): 285 - 295. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
