Tomotherapy: a new concept for the delivery of dynamic conformal radiotherapy

Beam characteristics and clinical possibilities of a new compact treatment unit design combining narrow pencil beam scanning and segmental multileaf collimation

Not until the last decade has flexible intensity modulated three-dimensional dose delivery techniques with photon beams become a clinical reality, first in the form of heavy metal transmission blocks and other beam compensators, then in dynamic and segmented multileaf collimation, and most recently by scanning high-energy narrow electron and photon beams. The merits of various treatment unit and bremsstrahlung target designs for high-energy photon therapy are investigated theoretically for two clinically relevant target sites, a cervix and a larynx cancer both in late stages. With an optimized bremsstrahlung target it is possible to generate photon beams with a half-width of about 3 cm at a source to axis distance (SAD) of 100 cm and an initial electron energy of 50 MeV. By making a more compact treatment head and shortening the SAD, it is possible to reduce the half-width even further to about 2 cm at a SAD of 70 cm and still have sufficient clearance between the collimator head and the patient. One advantage of a reduced SAD is that the divergence of the beam for a given field size on the patient is increased, and thus the exit dose is lowered by as much as 1%/cm of the patient cross section. A second advantage of a reduced SAD is that the electron beam on the patient surface will be only about 8 mm wide and very suitable for precision spot beam scanning. It may also be possible to reduce the beamwidth further by increasing the electron energy up to about 60 MeV to get a photon beam of around 15 mm half-width and an electron beam as narrow as 5 mm. The compact machine will be more efficient and easy to work with, due to the small gantry and the reduced isocentric height. For a given target volume and optimally selected static multileaf collimator, it is no surprise that the narrowest possible scanned elementary bremsstrahlung beam generates the best possible treatment outcome. In fact, by delivering a few static field segments with individually optimized scan patterns, it is possible to combine the advantage of being able to fine tune the fluence distribution by the scanning system with the steeper dose gradients that can be delivered by a few static multileaf collimator segments. It is demonstrated that in most cases a few collimator segments are sufficient and often a single segment per beam portal may suffice when narrow scanned photon beams are employed, and they can be delivered sequentially with a negligible time delay. A further advantage is the increase of therapeutically useful photons and improved patient protection, since the pencil beam is only scanned where the leaf collimator is open. Consequently, some of the problems associated with dynamic multileaf collimation such as the tongue and groove and edge leakage effects are significantly reduced. Fast scanning beam techniques combined with good treatment verification systems allow interesting future possibilities to counteract patient and internal organ motions in real time.

Tomotherapeutic portal imaging for radiation treatment verification

In their tomotherapy concept Mackie and co-workers proposed not only a new technique for IMRT but also an appropriate and satisfactory method of treatment verification. This method allows both monitoring of the portal dose distribution and imaging of the patient anatomy during treatment by means of online CT. This would enable the detection of inaccuracies in dose delivery and patient set-up errors. In this paper results are presented showing that a single electronic portal imaging device (EPID) could deliver all data necessary to establish such a complete verification system for tomotherapy and even other IMRT techniques. Consequently it has to be shown that it is able to record both the low-intensity photon fluences encountered in tomographic imaging and the intense photon transmission of each treatment field. The detector under investigation is a video-based EPID, the BIS 710 (manufactured by Wellhöfer Dosimetrie, Schwarzenbruck, Germany). To examine the suitability of the BIS for CT at 6 MV beam quality, different phantoms were scanned and reconstructed. The agreement between a diamond detector and BIS responses is quantitative. Tomographic reconstruction of a complete set of these transmission profiles resulted in images which resolve 3 cm large objects having a (theoretical) contrast to water of less than 9%. Three millimetre objects with a 100% contrast are clearly visible. The BIS signal was shown to measure photon fluence distributions. The reconstructed images possess a spatial and contrast resolution sufficient for accurate imaging of the patient anatomy, needed for treatment verification in many clinical cases.

Target position variability throughout prostate radiotherapy

PURPOSE

To quantify the variability in prostate and seminal vesicle position during a course of external beam radiotherapy, and to measure the proportion of target variability due to setup error.

METHODS AND MATERIALS

Forty-four weekly planning computerized tomography (CT) studies were obtained on six patients undergoing radiotherapy for prostate cancer. All patients were scanned in the radiotherapy treatment position, supine with an empty bladder, with no immobilization device. All organs were outlined on 3-mm-thick axial CT images. Anterior and lateral beam's eye view digitally reconstructed radiographs and multiplanar reformatted images were generated. The position of the prostate and seminal vesicles relative to the isocenter location as set that day was recorded for each CT study. Target position relative to a bony landmark was measured to determine the relative contribution of setup error to the target position variability.

RESULTS

The seminal vesicle and prostate position variability was most significant in the anterior-posterior (AP) direction, followed by cranial-caudal (CC) and mediolateral (ML) directions. Setup error contributed significantly to the total target position variability. Rectal filling was associated with a trend to anterior movement of the prostate, whereas bladder filling was not associated with any trends. Although most deviations from the target position determined at the initial planning CT scan were within 10 mm, deviations as large as 15 mm and 19 mm were seen in the prostate and seminal vesicles respectively. Target position variations were evenly distributed around the initial target position for some patient studies, but unpredictable patterns were also seen. From a simulation based on the observed variability in target position, the AP, CC, and ML planning target volume (PTV) borders around the clinical target volume (CTV) required for target coverage with 95% certainty are 12.4 mm, 10.3 mm, and 5.6 mm respectively for the prostate and 13.8 mm, 8.6 mm, and 3.9 mm respectively for the seminal vesicles.

CONCLUSION

Target position variability is significant during prostate radiotherapy, requiring large PTV borders around the CTV. This target position variability may be potentially reduced by improving the setup accuracy.

Feasibility of megavoltage portal CT using an electronic portal imaging device (EPID) and a multi-level scheme algebraic reconstruction technique (MLS-ART)

Although electronic portal imaging devices (EPIDs) are efficient tools for radiation therapy verification, they only provide images of overlapped anatomic structures. We investigated using a fluorescent screen/CCD-based EPID, coupled with a novel multi-level scheme algebraic reconstruction technique (MLS-ART), for a feasibility study of portal computed tomography (CT) reconstructions. The CT images might be useful for radiation treatment planning and verification. We used an EPID, set it to work at the linear dynamic range and collimated 6 MV photons from a linear accelerator to a slit beam of 1 cm wide and 25 cm long. We performed scans under a total of approximately 200 monitor units (MUs) for several phantoms in which we varied the number of projections and MUs per projection. The reconstructed images demonstrated that using the new MLS-ART technique megavoltage portal CT with a total of 200 MUs can achieve a contrast detectibility of approximately 2.5% (object size 5 mm x 5 mm) and a spatial resolution of 2.5 mm.

The impact of treatment complexity and computer-control delivery technology on treatment delivery errors

PURPOSE

To analyze treatment delivery errors for three-dimensional (3D) conformal therapy performed at various levels of treatment delivery automation and complexity, ranging from manual field setup to virtually complete computer-controlled treatment delivery using a computer-controlled conformal radiotherapy system (CCRS).

METHODS AND MATERIALS

All treatment delivery errors which occurred in our department during a 15-month period were analyzed. Approximately 34,000 treatment sessions (114,000 individual treatment segments [ports]) on four treatment machines were studied. All treatment delivery errors logged by treatment therapists or quality assurance reviews (152 in all) were analyzed. Machines "M1" and "M2" were operated in a standard manual setup mode, with no record and verify system (R/V). MLC machines "M3" and "M4" treated patients under the control of the CCRS system, which (1) downloads the treatment delivery plan from the planning system; (2) performs some (or all) of the machine set up and treatment delivery for each field; (3) monitors treatment delivery; (4) records all treatment parameters; and (5) notes exceptions to the electronically-prescribed plan. Complete external computer control is not available on M3; therefore, it uses as many CCRS features as possible, while M4 operates completely under CCRS control and performs semi-automated and automated multi-segment intensity modulated treatments. Analysis of treatment complexity was based on numbers of fields, individual segments, nonaxial and noncoplanar plans, multisegment intensity modulation, and pseudoisocentric treatments studied for a 6-month period (505 patients) concurrent with the period in which the delivery errors were obtained. Treatment delivery time was obtained from the computerized scheduling system (for manual treatments) or from CCRS system logs. Treatment therapists rotate among the machines; therefore, this analysis does not depend on fixed therapist staff on particular machines.

RESULTS

The overall reported error rate (all treatments, machines) was 0.13% per segment, or 0.44% per treatment session. The rate (per machine) depended on automation and plan complexity. The error rates per segment for machines M1 through M4 were 0.16%, 0.27%, 0.12%, 0.05%, respectively, while plan complexity increased from M1 up to machine M4. Machine M4 (the most complex plans and automation) had the lowest error rate. The error rate decreased with increasing automation in spite of increasing plan complexity, while for the manual machines, the error rate increased with complexity. Note that the real error rates on the two manual machines are likely to be higher than shown here (due to unnoticed and/or unreported errors), while (particularly on M4) virtually all random treatment delivery errors were noted by the CCRS system and related QA checks (including routine checks of machine and table readouts for each treatment). Treatment delivery times averaged from 14 min to 23 min per plan, and depended on the number of segments/plan, although this analysis is complicated by other factors.

CONCLUSION

Use of a sophisticated computer-controlled delivery system for routine patient treatments with complex 3D conformal plans has led to a decrease in treatment delivery errors, while at the same time allowing delivery of increasingly complex and sophisticated conformal plans with little increase in treatment time. With renewed vigilance for the possibility of systematic problems, it is clear that use of complete and integrated computer-controlled delivery systems can provide improvements in treatment delivery, since more complex plans can be delivered with fewer errors, and without increasing treatment time.

[Use of a multileaf collimator for the production of intensity-modulated beams]

In external radiotherapy, the use of intensity modulated fields has been proposed for tissue and non-homogeneity compensation or for the generation of conformal dose distributions. Multileaf collimators can be employed dynamically for the modulation of the X-ray field in two dimensions. Efficient dynamic collimation became possible due to advances in computer and linear accelerator technology. It presents a number of advantages over conventional methods such as the use of compensators. We have developed a program which calculates, from a given intensity distribution, the motion of the MLC leaves as a function of monitor units, and we have applied it on a Varian linear accelerator with a 40 pair multileaf collimator. The analysis of the experimental results demonstrates the feasibility and the potential of the method.

Registration of synthetic tomographic projection data sets using cross-correlation

Tomographic registration, a method that makes possible accurate patient registration directly from projection data, consists of three processing steps: (i) manual coarse positioning, (ii) tomographic projection set acquisition, and (iii) computer mediated refined positioning. In the coarse positioning stage, the degree of patient alignment is comparable with that achieved with the standard radiotherapy set-up. However, the accuracy requirements are somewhat more relaxed in that meticulous alignment of the patient using external laser indicators is not necessary. Instead, tomographic projection sets are compared with planning CTs in order to achieve improved patient set-up. The projection sets are cross-correlated to obtain the best-fit translation and rotation offsets. The algorithm has been tested on synthetic data with the incorporation of varying amounts of Gaussian pseudo-random noise. These tests demonstrate the algorithm's stability and also confirm that alignment can be achieved with an accuracy of less than one projection pixel.

Portal dose image (PDI) prediction for dosimetric treatment verification in radiotherapy. I. An algorithm for open beams

A method is presented for calculation of transmission functions for high energy photon beams through patients. These functions are being used in our clinic for prediction of portal dose images (PDIs) which are compared with PDIs measured with an electronic portal imaging device (EPID). The calculations are based on the planning CT-scan of the patient and on the irradiation geometry as determined in the treatment planning process. For each beam quality, the required input data for the algorithm for transmission prediction are derived from a limited number of measured beam data. The method has been tested for a PDI-plane at 160 cm from the focus, in agreement with the fixed focus-to-detector distance of our fluoroscopic EPIDs. For 6, 23 and 25 MV photon beams good agreement (approximately 1%) has been found between calculated and measured transmissions through anthropomorphic phantoms.

Dosimetric verification of the dynamic intensity-modulated radiation therapy of 92 patients

PURPOSE

To verify that optimized dose distributions provided by an intensity-modulated radiation therapy (IMRT) system are delivered accurately to human patients.

METHODS AND MATERIALS

Anthropomorphic phantoms are used to measure IMRT doses. Four types of verification are developed for: I) system commissioning with beams optimized to irradiate simulated targets in phantoms, II) plans with patient-optimized beams directed to phantoms simulating the patient, III) patient-phantom hybrid plans with patient-optimized beams calculated in phantom without further optimization, and IV) in vivo measurements. Phantoms containing dosimeters are irradiated with patient-optimized beams. Films are scanned and data were analyzed with software. Percent difference between verified and planned maximum target doses is defined as "dose discrepancy" (deltavp). The frequency distribution of type II deltavp from 204 verification films of 92 IMRT patients is fit to a Gaussian. Measurements made in vivo yield discrepancies specified as deltaivp, also fit to a Gaussian.

RESULTS AND DISCUSSION

Verification methods revealed three systematic errors in plans that were corrected prior to treatment. Values of [deltavp] for verification type I are <2%. Type II verification discrepancies are characterized by a Gaussian fit with a peak 0.2% from the centroid, and 158 [deltavp] <5%. The 46 values of [deltavp] >5% arise from differences between phantom and patient geometry, and from simulation, calculation, and other errors. Values of [deltavp] for verification III are less than half of the values of [deltavp] for verification II. A Gaussian fit of deltaivp from verification IV shows more discrepancy than the fit of deltavp, attributed to dose gradients in detectors, and exacerbated by immobilization uncertainty.

CONCLUSIONS

Dosimetric verification is a critical step in the quality assurance (QA) of IMRT. Hybrid Verification III is suggested as a preliminary quality standard for IMRT.

Radiotherapy for pediatric brain tumors

Brain tumors in children are a diverse group of diseases that require multidisciplinary and subspecialty expertise. Radiation therapy is an established treatment cornerstone for these pediatric tumors. Basic concepts of radiation biology and physics provide a framework for understanding the ongoing evolution in radiation delivery techniques and current treatment paradigms. Standard techniques of pediatric central nervous system radiotherapy are included in this review, as well as newer techniques including conformal therapy, stereotactic radiosurgery, and fractionated stereotactic radiotherapy. Examples are provided to illustrate differences in treatment approaches. The appropriate application of each technique is discussed, and then outcomes and treatment sequelae are compared.

Possibilities for tailoring dose distributions through the manipulation of electron beam characteristics

The influence of the properties of an electron beam on resulting dose distributions, and the potential benefits for dose conformity and optimizing dose distribution characteristics by electron beam manipulation, are theoretically examined. A simulated annealing routine is used to weight electron pencil beams of discrete energies incident at discrete locations and angles on one side of a phantom. The resulting optimal electron phase space provides a dose distribution which most closely approaches a desired distribution on the basis of a physical comparison. For simple desired distributions, intuitive results are obtained such as the benefits of energy modulation for distributing dose with depth, of angular and spatial modulation for overcoming disequilibrium effects and their combination in boosting surface doses. For a complex desired dose distribution, the optimization routine instigates a complex interplay of energy, angular and spatial modulation in attempting to achieve dose conformity. A significant result shows that, for a suitably selected beam energy, angular modulation can compensate for the variation in the depth of the distal edge of a superficial target. The effects of varying just energy for normally incident electrons are compared with those of varying the distribution of incidence angle (for monoenergetic electrons) and the combination of both, indicating the relative merits of the manipulation of available degrees of freedom.

Head scatter modelling for irregular field shaping and beam intensity modulation

Scattered radiation from within the treatment head can contribute significant dose to all parts of a radiotherapy treatment field. A multileaf collimator may be used to create an arbitrarily shaped field, and may also be used, under dynamic control, to modulate the beam intensity over the field. This method of intensity modulation is effectively a superposition of a large number of fields which have the same beam direction, but different shapes, and some of these shapes may have unusually small dimensions, particularly in the direction of the leaf movement. Two models for predicting the head scatter under these conditions have been investigated. These are a first-order Compton scatter approximation from the flattening filter, and an empirical fit to measured data using an exponential function. The first model only considers scatter from the flattening filter and has been applied to field sizes between 2 cm by 2 cm and 10 cm by 10 cm, where agreements are all within 1%. However it is not satisfactory at larger field sizes where small scatter contributions, from scattering sources other than the flattening filter, are integrated over large areas. The second model uses measured data between 4 cm by 4 cm and 30 cm by 30 cm to optimize the exponential function and is used to calculate the head scatter contribution for all field sizes. In this case good agreement is achieved over the full field size range, and hence this is a more generally applicable model. Results are presented for static irregularly shaped fields and intensity modulated beams created using a Philips multileaf collimator.

Initial experience with intensity-modulated conformal radiation therapy for treatment of the head and neck region

PURPOSE

The efficacy of a conventional, noninvasive fixation technique in combination with a commercially available system for conformal radiotherapy by intensity modulation of the treatment beam has been studied.

METHODS AND MATERIALS

A slice-by-slice arc-rotation approach was used to deliver a conformal dose to the target and patient fixation was performed by means of thermoplastic casts. Eleven patients have been treated, of which 9 were for tumors of the head and neck region and 2 were for intracranial lesions. A procedure for target localization and verification of patient positioning suitable for this particular treatment technique has been developed based on the superposition of digitized portals with plots generated from the treatment-planning system. A dosimetric verification of the treatment procedure was performed with an anthropomorphic phantom: both absolute dose measurements (alanine and thermoluminescent detectors) and relative dose distribution measurements (film dosimetry) have been applied. The dose delivered outside the target has also been investigated.

RESULTS

The dose verification with the anthropomorphic phantom yielded a ratio between measured and predicted dose values of 1.0 for different treatment schedules and the calculated dose distribution agreed with the measured dose distribution. Day-to-day variations in patient setup of 0.3 cm (translations) and 2.0 degrees (rotations) were considered acceptable for this particular patient population, whereas the verification protocol allowed detection of 0.1 cm translational errors and 1.0 rotational errors.

CONCLUSIONS

The noninvasive fixation technique in combination with an adapted verification protocol proved to be acceptable for conformal treatment of the head and neck region. Dose measurements, in turn, confirmed the predicted dose values to the target and organs at risk within uncertainty. Daily monitoring becomes mandatory if an accuracy superior to 0.1 cm and 1.0 degree is required for patient setup.

Modulated beam conformal therapy for head and neck tumors

PURPOSE

The goal of modulated-beam conformal therapy is to reduce the dose to healthy tissue and sensitive structures around a uniformly irradiated target volume. Multiple intensity-modulated fields offer improved tissue-sparing dose distributions. New computer-based systems for planning and delivering such treatments may soon be available from different commercial sources that will make the formulation of an intensity-modulated treatment plan and its execution widely available at any treatment facility that has the resources to acquire the necessary equipment. This work reports on a study of the integration of two such systems.

METHODS AND MATERIALS

Treatment planning was done using a commercially available inverse planning algorithm based on simulated annealing. The plans arbitrarily assumed nine coplanar x-ray beams at nonopposed gantry angles. Intensity modulation was computed for each beam. The modulated field at each gantry angle was broken down into a series of uniform (nonmodulated) subfields, which could be delivered as a sequence to produce the desired dose distribution. Because a large number of subfields was delivered, a multileaf collimator (MLC) was used for field shaping. This allowed rapid and accurate field shaping for treatments made up of several hundred subfields. Computer control of the MLC and linear accelerator allowed delivery of doses less than .01 Gy per subfield. Treatment was delivered on a prototype, computer-controlled accelerator and MLC system. Resulting dose distributions were analyzed using film and an anatomically specific, homogeneous phantom.

RESULTS

The treatment plans were evaluated using dose-volume histogram analysis. The plans provided acceptably uniform irradiation of the target volume without exceeding dose tolerances for nearby critical structures. The plans were successfully delivered by a prototype dynamic MLC. The time needed to deliver a sequence of subfields at one gantry angle ranged from 0.7 to 2.0 min. Isodoses from film agreed reasonably well with planned isodose distributions.

CONCLUSIONS

It is feasible to plan and deliver fixed gantry, modulated-beam conformal therapy for head and neck tumors with systems being developed commercially. The planned dose distributions exhibit significant potential for sparing closely spaced normal tissue structures in the head and neck.

The influence of angular misalignment on fixed-portal intensity modulated radiation therapy

A method has been developed to estimate potential dose errors due to linear accelerator angular setting misalignments of Intensity Modulated Radiation Therapy (IMRT) treatments. A first-order approximation to the dose error at a point is modeled as the dot product of the dose gradient and the shift vector of the point due to the rotational error. The analysis method is applied to a previously published set of optimized fluences for a 50 MV IMRT pelvis irradiation. Three of the published cases exhibiting a wide range of modulation are presented; a rectangular open field, a field optimized for a static multileaf collimator defining the portal outline coupled with a single broad bremsstrahlung profile modulation, and a fully modulated field using a physical modulator. To examine the energy dependence of angle setting errors, the study is repeated using the same fluence distributions, but with a dose-spread kernel appropriate for a 6 MV photon beam. The collimator angle error is set to 2 degree, and the dose error determined with both a centrally located isocenter and an isocenter chosen to model a split-field geometry. The dose error due to a 2 degree gantry setting error is assessed at a plane 10 cm distal to the isocenter. The mathematical form of the dose error due to couch motion is similar to the other two errors, so the dose error resulting from a couch angle missetting is not presented. The magnitude of the errors is largest for the 6 MV beam, while the volume encompassed by the errors is greater for the 50 MV beam. The gantry error yields the largest dose error values, with the 6 MV modulated case presenting dose errors of greater than 40%.

Estimates of whole-body dose equivalent produced by beam intensity modulated conformal therapy

PURPOSE

To estimate the dose delivered to patients by photons and neutrons outside the radiation fields when beam intensity modulation conformal radiotherapy is given. These estimates are then used to compute the risk of secondary cancers as a sequela of the radiation therapy.

MATERIALS AND METHODS

The x-ray and neutron leakage accompanying two beam-intensity modulation techniques delivered by currently available linear accelerators were estimated for 6-MV, 18-MV, and 25-MV x-ray energies. Estimates of whole-body dose equivalents were determined using leakage measurements reported in the literature and treatment parameters derived for two modulated beam-intensity conformal therapy techniques. Risk values recommended by the National Council on Radiation Protection and Measurements (NCRP) were used to estimate the resulting risk of fatal radiation-induced cancer for 70.00 Gy prescribed tumor doses.

RESULTS

The computed worst-case risks of secondary cancers increased in the range from 1.00% for 6-MV x-rays to 24.4% for 25-MV x-rays.

CONCLUSIONS

Careful consideration should be made of the risks associated with secondary whole-body radiation before implementation of beam intensity modulated conformal therapy at x-ray energies greater than 10 MV.

Inverse radiotherapy planning for a concave-convex PTV in cervical and upper mediastinal regions. Simulation of radiotherapy using an Alderson-RANDO phantom. Planning target volume

AIM

Three-dimensional inverse treatment planning with modulated beams was applied for dosimetric optimization of a lengthy (22 cm) and complex (concave-convex) shaped planning target volume (PTV) in the cervical and upper mediastinal regions.

MATERIAL AND METHOD

The planning was done for 9 coplanar beams spaced evenly at 40 intervals. Properties of 15 MV photons from a linear accelerator were simulated. The optimization of the fluence modulation profiles for each beam was based on a definition of the desired/permitted relative dose levels in the PTV and organs at risk, and a definition of the strengths of the constraints to achieve these objectives.

RESULTS

An adequate dose delivery to the PTV and protection of the spinal cord are completely achievable. The dose delivered to the lungs is clinically acceptable with respect to the risk of radiation-induced pneumonitis. For reasons of physics, no further decrease in the radiation burden on the lungs can be attained with X-rays without compromising the PTV coverage. The radiation burden on some critical part of normal tissues was effectively reduced by application of a dummy organ at risk.

CONCLUSION

The inverse planning is an effective method for conformal radiotherapy of large tumors as well. However, the power of the technique is insufficient when the tolerance dose of the neighbouring normal tissue is too low and its volume effect is high. Although requiring further operator interactions, introduction of dummy organs at risk may be of help in reducing the radiation burden on normal tissues.

An investigation of tomotherapy beam delivery

Experimental simulations for tomotherapy beam delivery were performed using a computer-controlled phantom positioner, a cylindrical phantom, and a 6 MV x-ray slit beam. Both continuous helical beam and sequential segmented tomotherapy (SST) beam deliveries were evaluated. Beam junctioning problem due to couch indexing error or field width errors presented severe dose uniformity perturbations for SST, while the problem was minimized for helical beam delivery. Longitudinal breathing motions were experimentally simulated for helical and SST beam delivery. While motions reduced the dose uniformity perturbations for SST, small artifacts in dose uniformity can be introduced for helical beam delivery. With typical breath frequency and magnitude, for a slit beam of 2.0 cm width at 4 rpm, the dose uniformity perturbation was not significant. A running start/stop technique was implemented with helical beam delivery to sharpen the 20%-80% longitudinal dose fall-off from 1.5 to 0.5 cm. The latter was comparable to the corresponding dose penumbra of a conventional 6 MV 10 x 10 cm2 field. All together, helical beam delivery showed advantages over SST for tomotherapy beam delivery under similar delivery conditions.

Optimization of relative weights and wedge angles in treatment planning

An efficient technique to optimize beam weights and wedge angles in radiotherapy treatment planning has been developed. Based on the fact that a wedged field can be regarded as a superposition of an open field and a nominal wedged field, this approach reduces the problem of finding J beam weights and the corresponding wedge angles to optimizing a linear system with 2J unknowns (weights of J open beams and J nominal wedged beams), where J is the total number of incident beam directions. Two iterative algorithms similar to the iterative-least-square technique in image reconstruction are used to optimize the system. Application of the algorithms to two specific examples shows that this technique can reduce treatment planning time and effort, and promises to create a better solution for an arbitrarily complex treatment plan.

Optimizing the movement of a single absorber for 1D non-uniform dose delivery by (fast) simulated annealing

A new simplified technique for 1D non-uniform dose delivery using a single dynamic absorber, driven by a computer system, has been recently proposed together with a simple analytic algorithm. This technique uses an optimized 'stepped' absorber's speed profile and the generated fluence profile is an approximation of the desired radiation beam. In the case of non-uniform beam profiles with multiple maxima/minima, the original proposed 'stepping algorithm' has some limitations and produces a too rough approximation of the desired profiles. In order to increase the agreement between desired and generated profiles, more sophisticated optimization schemes are required. In this paper we have applied a variant of simulated annealing (SA) as a statistical optimization algorithm to further investigate the possibilities and the limits of the single-absorber technique in the field of 1D intensity modulation. In the current application the cost function used is the mean square root of the percentage differences between desired and generated profiles, the absorber's resting times have been chosen as optimization variables and at each iteration just one variable is randomly changed, adding an incremental 'grain'. A Cauchy generating function is used, different cooling schedules are evaluated; constraints related to our apparatus are introduced and starting annealing parameters are set after some initial optimization tests. The method is tested in reproducing theoretical non-uniform beams, by comparing desired modulated fluence profiles with calculated fluence profiles obtainable with the single absorber after the derivation of optimized speed profiles by the proposed SA approach. The results of these simulations show that the application of the SA method optimizes the single absorber's performance and that clinically important modulated beams useful for conformal radiotherapy can be accurately reproduced.

Evaluation of dose calculation algorithm of the peacock system for multileaf intensity modulation collimator

PURPOSE

To evaluate the dose calculation algorithm used in the inverse treatment planning computer system for the intensity modulation multileaf collimator.

METHODS AND MATERIALS

The inverse treatment-planning computer system calculates the intensities of multiple pencil beams to achieve an optimal distribution and modulates the beam intensity through the special multileaf collimator. The system's dose calculation algorithm made the two basic assumptions: (a) The tissue-maximum ratios (TMRs) of a single pencil beam have the same values as TMRs for raylines through each pencil beam that are determined from percentage depth dose isodose curves along the long axis of the 2 x 20 cm2 field with all leaves open; and (b) the relative output factors (ROF) of each pencil beam also have the same values as the rayline TMR at d(max) of the 2 x 20 cm2 field. To verify these two assumptions, a special multileaf collimator was installed to our linear accelerator which produces 4 MV x-rays. The TMRs and ROFs for the single leaves 1 through 10 were measured using an ion chamber and TLD dosimeter in either a water or a polystyrene phantom. The values of rayline TMRs were calculated from the measured crossplane isodose curves of the 2 x 20 cm2 field. Comparisons were made between these two sets of data.

RESULTS

Based on our measurements, we found that the ROFs of a pencil beam obtained from the rayline TMRs at d(max) are as much as 7.6% greater than that of single pencil beams. The ROF of the 1 x 1 cm2 pencil beam is 4 and 6.5% less than that of a cluster of four neighboring pencil beams forming a 2 x 2 cm2, and a 2 x 20 cm2 field respectively. However, the rayline TMRs are generally larger than the TMRs of a single pencil beam. At a depth of 8 cm, the average depth in the middle of intracranial space, the rayline TMRs of the pencil beams of leaves 1 and 10 are 5.4 and 9% higher than a single pencil beam TMR at the same depth, respectively. Also interesting is to note that the TMRs of each of the single pencil beams were found to be equal.

CONCLUSIONS

In our article, evaluations and comparisons of TMRs and ROFs were made for two extreme conditions. The measured values of TMRs and ROFs of a single beam have been shown to be significantly different from those used in the calculations. Because both the TMR and ROF are influenced by the scattering radiation in the same direction, the deviations for these two factors would be expected to be magnified. Thus, for the two extreme situations we have investigated, dose deviations would be on the order of 15%. In real patient treatment; of course, these deviations may be somewhat less, but still significant. Our results, however, show that further investigations are warranted.

A dual computed tomography linear accelerator unit for stereotactic radiation therapy: a new approach without cranially fixated stereotactic frames

PURPOSE

To perform stereotactic radiation therapy (SRT) without cranially fixated stereotactic frames, we developed a dual computed tomography (CT) linear accelerator (linac) treatment unit.

METHODS AND MATERIALS

This unit is composed of a linac, CT, and motorized table. The linac and CT are set up at opposite ends of the table, which is suitable for both machines. The gantry axis of the linac is coaxial with that of the CT scanner. Thus, the center of the target detected with the CT can be matched easily with the gantry axis of the linac by rotating the table. Positioning is confirmed with the CT for each treatment session. Positioning and treatment errors with this unit were examined by phantom studies. Between August and December 1994, 8 patients with 11 lesions of primary or metastatic brain tumors received SRT with this unit. All lesions were treated with 24 Gy in three fractions to 30 Gy in 10 fractions to the 80% isodose line, with or without conventional external beam radiation therapy.

RESULTS

Phantom studies revealed that treatment errors with this unit were within 1 mm after careful positioning. The position was easily maintained using two tiny metallic balls as vertical and horizontal marks. Motion of patients was negligible using a conventional heat-flexible head mold and dental impression. The overall time for a multiple noncoplanar arcs treatment for a single isocenter was less than 1 h on the initial treatment day and usually less than 20 min on subsequent days. Treatment was outpatient-based and well tolerated with no acute toxicities. Satisfactory responses have been documented.

CONCLUSION

Using this treatment unit, multiple fractionated SRT is performed easily and precisely without cranially fixated stereotactic frames.

The field-matching problem as it applies to the peacock three dimensional conformal system for intensity modulation

PURPOSE

Intensity modulated beam systems have been developed as a means of creating a high-dose region that closely conforms to the prescribed target volume while also providing specific sparing of organs at risk within complex treatment geometries. The slice-by-slice treatment paradigm used by one such system for delivering intensity modulated fields introduces regions of dose nonuniformity where each pair of treatment slices abut. A study was designed to evaluate whether or not the magnitude of the nonuniformity that results from this segmental delivery paradigm is significant relative to the overall dose nonuniformity present in the intensity modulation technique itself. An assessment was also made as to the increase in nonuniformity that would result if errors were made in indexing during treatment delivery.

METHODS AND MATERIALS

Treatment plans were generated to simulate correctly indexed and incorrectly indexed treatments of 4, 10, and 18 cm diameter targets. Indexing errors of from 0.1 to 2.0 mm were studied. Treatment plans were also generated for targets of the same diameter but of lengths that did not require indexing of the treatment couch.

RESULTS

The nonuniformity that results from the intensity modulation delivery paradigm is 11-16% for targets where indexing is not required. Correct indexing of the couch adds an additional 1-2% in nonuniformity. However, a couch indexing error of as little as 1 mm can increase the total nonuniformity to as much as 25%. All increases in nonuniformity from indexing are essentially independent of target diameter.

CONCLUSIONS

The dose nonuniformity introduced by the segmental strip delivery paradigm is small relative to the nonuniformity present in the intensity modulation paradigm itself. A positioning accuracy of better than 0.5 mm appears to be required when implementing segmental intensity modulated treatment plans.

Intensity-modulated arc therapy with dynamic multileaf collimation: an alternative to tomotherapy

The desire to improve local tumour control and cure more cancer patients, coupled with advances in computer technology and linear accelerator design, has spurred the developments of three-dimensional conformal radiotherapy techniques. Optimized treatment plans, aiming to deliver high dose to the target while minimizing dose to the surrounding tissues, can be delivered with multiple fields each with spatially modulated beam intensities or with multiple-slice treatments. This paper introduces a new method, intensity-modulated arc therapy (IMAT), for delivering optimized treatment plans to improve the therapeutic ratio. It utilizes continuous gantry motion as in conventional arc therapy. Unlike conventional arc therapy, the field shape, which is conformed with the multileaf collimator, changes during gantry rotation. Arbitrary two-dimensional beam intensify distributions at different beam angles are delivered with multiple superimposing arcs. A system capable of delivering the IMAT has been implemented. An example is given that illustrates the feasibility of this new method. Advantages of this new technique over tomotherapy and other slice-based delivery schemes are also discussed.

Which is the most suitable number of photon beam portals in coplanar radiation therapy?

PURPOSE

Computer-controlled milling machines for compensator manufacture, dynamic multileaf collimators, and narrow scanned electron or bremsstrahlung photon beams have opened up new possibilities to shape nonuniform fluence profiles and have thus, paved the road for truly three dimensional (3D) dose delivery. The present paper investigates the number of beam portals required to optimize coplanar radiation therapy using uniform and nonuniform dose delivery.

METHODS AND MATERIALS

A recently developed algorithm has been used for optimization of the dose delivery in such a way that the probability of achieving tumor control without causing severe reactions in healthy normal tissues becomes as high as possible. This method has been used to optimize the delivered dose distribution for an increasing number of beam portals to determine how many coplanar beam portals are desirable to safely achieve a good treatment outcome. Target volumes in the head and neck, thorax, and abdomen have been investigated.

RESULTS

Nonuniform dose delivery allows a considerable improvement in the treatment outcome compared to uniform dose delivery. This is evident both from the probability of achieving complication-free tumor control and the value of relevant properties of the dose distribution, such as the mean value and the standard deviation of the mean dose to target volume and organs at risk. The results also show a close relationship between the dose distribution parameters and the probability of achieving complication-free tumor control. The level of complication-free tumor control first increases rapidly when the number of beam portals is increased, but already reaches a level of saturation after three to five beam portals. When the saturation level has been reached, the standard deviation of the mean dose to the target volume is around 3%.

CONCLUSIONS

To achieve optimal expectation value of the treatment outcome, within an accuracy of a few percent as measured by the probability of achieving complication-free tumor control, it is generally sufficient to use three nonuniform beam portals. A very large number of coplanar beams may only raise the probability of achieving complication-free tumor control by 1 to 2%. However, good treatment outcome with three beam portals requires that the directions of incidence of the coplanar nonuniform beams are optimally selected. If, on the other hand, the treatment is performed using uniform beams, it is not possible, even with an infinite number of fields, to obtain as high a level of complication-free tumor control as with a few nonuniform beams. From an optimization point of view, it is sufficient to reach a relative standard deviation of the mean dose to the target volume of around 3%. Improved dose homogeneity beyond this level will, in general, not significantly improve the complication-free tumor control.

An iterative filtered backprojection inverse treatment planning algorithm for tomotherapy

PURPOSE

An inverse treatment planning algorithm for tomotherapy is described.

METHODS AND MATERIALS

The algorithm iteratively computes a set of nonnegative beam intensity profiles that minimizes the least-square residual dose defined in the target and selected normal tissue regions of interest. At each iteration the residual dose distribution is transformed into a set of residual beam profiles using an inversion method derived from filtered backprojection image reconstruction theory. These "residual" profiles are used to correct the current beam profile estimates resulting in new profile estimates. Adaptive filtering is incorporated into the inversion model so that the gross structure of the dose distribution is optimized during initial iterations of the algorithm, and the fine structure corresponding to edges is obtained at later iterations. A three dimensional, kernel based, convolutions/superposition dose model is used to compute dose during each iteration.

RESULTS

Two clinically relevant treatment planning examples are presented illustrating the use of the algorithm for planning conformal radiotherapy of the breast and the prostate. Solutions are generally achieved in 10-20 iterations requiring about 20 h of CPU time using a midrange workstation. The majority of the calculation time is spent on the three-dimensional dose calculation.

CONCLUSIONS

The inverse treatment planning algorithm is a useful research tool for exploring the potential of tomotherapy for conformal radiotherapy. Further work is needed to (a) achieve clinically acceptable computation times; (b) verify the algorithm using multileaf collimator technology; and (c) extend the method to biological objectives.

Treatment delivery accuracy in intensity-modulated conformal radiotherapy

The importance of geometric and dosimetric inaccuracies in the delivery of conformal radiotherapy using intensity-modulated x-ray fields is assessed. It is shown that although the incident intensity or fluence profiles may have large gradients, the resulting absorbed dose profile gradients are shallower because of electron transport and photon scatter. This makes the interleaving of dose contributions from the many intensity-modulated fields used less critically dependent on their relative positions and on the accuracy of profile generation. It is found that the expected geometric and dosimetric inaccuracies do not make this technique impracticable provided that good patient fixation and beam delivery can be assured.

Interventional telemedicine for noninvasive neuroradiosurgery: remote-site high-performance computing, mathematical optimization, and virtual scenario simulation

Interventional Telemedicine may have potential utility in providing connectivity and access to specialized high performance computing and advanced software resources in support of clinical procedures in the field of Minimally-Invasive Surgery and Stereotactic Linear Accelerator (LINAC) Radiosurgery. Such interventions may benefit from application of nonlinear quadratic inverse solution methods designed to provide the capability to reverse optimize a 'best case' treatment plan. The formidable decision-making challenges posed by increasingly complex optimized data and progressively versatile LINAC delivery systems require volume visualization of projected treatment data and imaging anatomy via photorealistic rendering and virtual scenario simulation techniques. Both these new directions are heavily dependent on access to specialized high performance computing platforms solely accessible via broad-bandwidth network connectivity. This pilot project presents resimulation of retrospective radiosurgical case data using inverse solution optimization models running on workstation clusters and then volume rendered and simulated on the Princeton Graphics Engine Supercomputer. Evidence for effective utilization of such optimization and virtual simulation methods running on remotely accessed, distant high performance computing resources is discussed in view of the potential for long-term clinical investigation and eventual development of Interventional Telemedicine as a clinically practical approach for providing support to remote or non-urban radiosurgery centers in the industrialized and developing world.

Dynamic beam modulation by using a single computer-controlled absorber

The authors have developed an apparatus able to generate ID intensity-modulated beams, using only one moving absorber within the irradiation field. A procedure for deriving optimized absorber-speed profiles in order to produce the desired fluence/dose profiles has been suggested. Experimental tests show that the system should be sufficiently reliable in reproducing modulated beam profiles of different shape: expected relative doses against measured relative doses have been found to be in agreement in a number of situations within 3% using a nonfocused device. A better agreement should be expected using a focused apparatus (which is currently being developed). Beam modulation by single absorber cannot modulate the beam fluence in any was one wishes, due to physical constraints, which depend on the absorber and field widths and on the shape of the desired fluence profile. However, the authors show that this simple and low-cost tool could offer, with a sufficient degree of accuracy, the possibility of modulating the beam fluence with a high degree of versatility. In particular, a procedure or performing tissue-missing compensation by single-absorber dynamic beam modulation is suggested. Moreover, 'strongly' modulated beam profiles can be created, showing that this simple technique could also have some interesting applications in the field of conformal radiotherapy by non-uniform dose delivery.

Dynamic X-ray compensation for conformal radiotherapy by means of multi-leaf collimation

The application of a multiple fixed field technique employing individually shaped and intensity-modulated beams makes it possible to produce dose distributions of high conformity even in the case of concave target volumes. With the technique presented here arbitrary intensity-modulated beams for the practical solution of the inverse problem can be generated. It is also possible to omit wedges, blocks and compensators in conventional radiotherapy. A continuous unidirectional sweep of independently computer-controlled leaves of a multi-leaf collimator is used to modulate the primary uniform beam. A new algorithm is introduced that computes the leaf trajectories. Also, a method is presented that accounts for leaf penumbra and transmission, which causes the generated fluence distribution to deviate from the desired fluence distribution. An optimization algorithm minimizing this deviation is described. The algorithm calculating the leaf trajectories, as well as the method considering penumbra and transmission and the successive optimization technique are used to calculate examples. Treatment times are calculated and compared to those needed when using compensators. A relationship between the treatment time and the maximum leaf speed is also deduced. To achieve good performance the maximum leaf speed should be no less than 20 mm/s.