Investigation of dosimetric differences between the TMR 10 and convolution algorithm for Gamma Knife stereotactic radiosurgery

Deep Learning-Based Heterogeneity Correction of the Homogeneous Dose Distribution for Single Brain Tumors in Gamma Knife Radiosurgery

Purpose

Heterogeneity correction is vital in radiation therapy treatment planning to ensure accurate dose delivery. Brain cancer stereotactic treatments, like Gamma Knife radiosurgery (GKRS), often rely on homogeneous water-based calculations despite the potential heterogeneity impact near bony structures. This study aims to develop a method for generating synthetic dose plans incorporating heterogeneity effects without additional computed tomography (CT) scans.

Methods and Materials

Magnetic resonance imaging and CT images, TMR10-based, and convolution-based dose distributions were used from 100 retrospectively collected and 22 prospectively collected GKRS patients. A conditional Generative Adversarial Network was trained to translate TMR10 into synthetic convolution (sConv) doses.

Results

The generated sConv dose demonstrated qualitative and quantitative similarity to the actual convolution (Conv) dose, showcasing better agreement of dose distributions and improved isodose volume similarity with the Conv dose in comparison to the TMR10 dose (γ pass rate; sConv dose, 92.43%; TMR10 dose, 74.18%. Prescription isodose dice; sConv dose, 91.7%; TMR10 dose, 89.7%). Skull-induced scatter and attenuation effects were accurately reflected in the sConv dose, indicating the usefulness of the new dose prediction model as an alternative to the time-consuming convolution dose calculations.

Conclusions

Our deep learning approach offers a feasible solution for heterogeneity-corrected dose planning in GKRS, circumventing additional CT scans and lengthy calculation times. This method's effectiveness in preserving dose distribution characteristics in a heterogeneous medium while only requiring a homogeneous dose plan highlights its utility for including the process in the routine treatment planning workflows. Further refinement and validation with diverse patient cohorts can enhance its applicability and impact in clinical settings.

Adaptation of dose-prescription for vestibular schwannoma radiosurgery taking body contouring method and heterogeneous material into account

BACKGROUND

Majority of vestibular schwannoma (VS) patients have undergone gamma-knife radiosurgery (GKRS) with favorable results. Clinical evidence is derived from doses calculated with a type-a algorithm, which in this case assumes all material to be water. A type-b algorithm (Convolution algorithm [CA]) taking tissue heterogeneity into account is available. Historically, body contour is defined using a 16-point approximation, whereas modern softwares generate the body from Magnetic Resonance Imaging (MRI). The accuracy in dose-calculation algorithms (DCA) and contouring method (CM) will have a significant influence in the relation between clinical outcome and dosimetric data. The objective was to investigate the impact of DCA and CMs on dose distribution while preserving treatment conditions.

METHODS

Treatment plans for 16 VS patients were recalculated in terms of DCA and CM. The difference in the dose covering 99% of the VS (DVS99%) depending on CM and DCA was estimated. The difference in DVS99% was used to adopt the prescription of new CA-based plans. CA-plans were recalculated to TMR10 to evaluate clinical treatability, as clinical evidence is derived from TMR10-doses.

RESULTS

Both CM and DCA had a significant impact on the dose to VS and surrounding structures. CM altered the doses homogenously by 2.1-3.3%, whereas DCA heterogeneously by 5.0-10.7%. An increase of 9.1[8.1, 10.0]% was found for DVS99% and the CA-plans recalculated into TMR10 resulted in clinically treatable plans.

INTERPRETATION

We conclude that transferring to more modern algorithms that take tissue heterogeneity into account heterogeneously alter dose distributions. This work establishes a safe pathway to adopt prescription dose for VS while preserving clinical treatability.

ISRS Technical Guidelines for Stereotactic Radiosurgery: Treatment of Small Brain Metastases (≤1 cm in Diameter)

PURPOSE

The objective of this literature review was to develop International Stereotactic Radiosurgery Society (ISRS) consensus technical guidelines for the treatment of small, ≤1 cm in maximal diameter, intracranial metastases with stereotactic radiosurgery. Although different stereotactic radiosurgery technologies are available, most of them have similar treatment workflows and common technical challenges that are described.

METHODS AND MATERIALS

A systematic review of the literature published between 2009 and 2020 was performed in Pubmed using the Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA) methodology. The search terms were limited to those related to radiosurgery of brain metastases and to publications in the English language.

RESULTS

From 484 collected abstract 37 articles were included into the detailed review and bibliographic analysis. An additional 44 papers were identified as relevant from a search of the references. The 81 papers, including additional 7 international guidelines, were deemed relevant to at least one of five areas that were considered paramount for this report. These areas of technical focus have been employed to structure these guidelines: imaging specifications, target volume delineation and localization practices, use of margins, treatment planning techniques, and patient positioning.

CONCLUSION

This systematic review has demonstrated that Stereotactic Radiosurgery (SRS) for small (1 cm) brain metastases can be safely performed on both Gamma Knife (GK) and CyberKnife (CK) as well as on modern LINACs, specifically tailored for radiosurgical procedures, However, considerable expertise and resources are required for a program based on the latest evidence for best practice.

Gamma Knife evolving instrumentation

There have been advances in both the hardware and software used in GKNS. The first major change in hardware had been Gamma Knife PERFEXION which introduced in 2006 had given more space for treatment, and removed the need for helmets, facilitating the treatment of complex conditions. Gamma Knife ICON was commissioned first in 2017. This has two important changes. It is based on the PERFEXION model, but it is constructed to permit frameless treatments. It also has an attached CBCT apparatus which may be used to define the stereotactic space. The Gamma Knife software has also improved in two important respects. The speedy calculations available to modern computer power has enabled improvements in the accuracy of the determination of intracranial radiation absorption between source and target. The other improvement has been the introduction of inverse treatment planning which continues to be under development.

Dosimetry

The dosimeters used to measure radiation dose produce a value which has to be calibrated to be in keeping with the values in an approved laboratory, which will be one of an international network of such laboratories at the center of which is the Bureau International des Poids et Mesures in France (BIPM). Dosimeters work by producing a quantitatively proportional change in status to the intensity of the radiation being measure. Amongst the techniques in use are thermoluminescent devices, radiographic film, radiochromic film, semiconductors, ionization chambers, silicon diodes and gel dosimeters. The Gamma Knife radiation has been difficult to measure directly because the beams have been to fine for accurate measurement by commonly available dosimeters. For more modern dosimeters this is less of a problem. During the treatment of a patient, a variety of indices are recorded to assist in the standardization and accuracy of treatment. Having determined the dose in the beams, it is necessary to calculate how much energy is lost during the passage of radiation from the source to the target. There has been a steady evolution of these calculations to make them more accurate.

Analysis of treatment planning parameters in the Gamma Knife® technique for different prescription isodoses and volumes of meningiomas

The following Indexes: Homogeneity, Gradient, Conformity, Paddick Conformity and New Conformity of the dose distribution were compared. The parameters to assess a high dose to the organs at risk: V₁₀/TV, V_90%/TV and the Integral Dose were discussed. The higher the prescription isodose, the more uniform the dose distribution in the target, which is highly beneficial in the case of larger tumor sizes due to the lower risk of complications. For smaller tumors, higher dose heterogeneity is desirable. This can be obtained with a 40% prescription isodose.

Impact of tissue heterogeneity correction on Gamma Knife stereotactic radiosurgery of acoustic neuromas

PURPOSE/OBJECTIVES

Treatment planning systems (TPS) for Gamma Knife stereotactic radiosurgery (GK-SRS) include TMR10 algorithms, which assumes tissue homogeneity equivalent to water, and collapsed-cone convolutional (CCC) algorithms, which accounts for tissue inhomogeneity. This study investigated dosimetric differences between TMR10 and CCC TPS for acoustic neuromas (ANs) treated with GK-SRS.

MATERIALS/METHODS

A retrospective review of 56 AN treated with GK-SRS was performed. All patients underwent MRI and CT imaging during their initial treatment and were planned using TMR10. Each plan was recalculated with CCC using electron density extracted from CT. Parameters of interest included Dmax, Dmin, D50%, cochlea Dmax, mean cochlea dose, target size, and laterality (>20 mm from central axis).

RESULTS

Median target volume of patients was 1.5 cc (0.3 cc-2.8 cc) with median dose of 12 Gy prescribed to the 50% isodose line. Compared to CCC algorithms, the TMR10 calculated dose was higher: Dmax was higher by an average 6.2% (p < 0.001), Dmin was higher by an average 3.1% (p < 0.032), D50% was higher by an average of 11.3%. For lateralized targets, calculated Dmax and D50% were higher by 7.1% (p < 0.001) and 10.6% (p < 0.001), respectively. For targets <1 cc, Dmax and D50% were higher by 8.9% (p ≤ 0.009) and 12.1% (p ≤ 0.001), respectively. Cochlea Dmax was higher, by an average of 20.1% (p < 0.001).

CONCLUSION

There was a statistically significant dosimetric differences observed between TMR10 and CCC algorithms for AN GK-SRS, particularly in small and lateralized ANs. It may be important to note these differences when relating GK-SRS with standard heterogeneity-corrected SRS regimens.

Impact of the skull contour definition on Leksell Gamma Knife® Icon™ radiosurgery treatment planning

INTRODUCTION

The Gamma Knife^® planning software (TMR 10, Elekta Instruments, AB, Sweden) affords two ways of defining the skull volume, the "historical" one using manual measurements (still perform in some centers) and the new one using image-based skull contours. Our objective was to assess the potential variation of the dose delivery calculation using consecutively in the same patients the two above-mentioned techniques.

MATERIALS AND METHODS

We included in this self-case-control study, 50 patients, treated with GKRS between July 2016 and January 2017 in Lausanne University Hospital, Switzerland, distributed among four groups: convexity targets (n = 18), deep-seated targets (n = 13), vestibular schwannomas (n = 11), and trigeminal neuralgias (n = 8). Each planning was performed consecutively with the 2 skull definition techniques. For each treatment, we recorded the beam-on time (min), target volume coverage (%), prescription isodose volume (cm³), and maximal dose (Gy) to the nearest organ at risk if relevant, according to each of the 2 skull definition techniques. The image-based contours were performed using CT scan segmentation, based upon a standardized windowing for all patients.

RESULTS

The median difference in beam-on time between manual measures and image-based contouring was + 0.45 min (IQR; 0.2-0.6) and was statistically significant (p < 0.0001), corresponding to an increase of 1.28% beam-on time per treatment, when using image-based contouring. The target location was not associated with beam-on time variation (p = 0.15). Regarding target volume coverage (p = 0.13), prescription isodose volume (p = 0.2), and maximal dose to organs at risk (p = 0.85), no statistical difference was reported between the two skull contour definition techniques.

CONCLUSION

The beam-on time significantly increased using image-based contouring, resulting in an increase of the total dose delivery per treatment with the new TMR 10 algorithm. Other dosimetric parameters did not differ significantly. This raises the question of other potential impacts. One is potential dose modulation that should be performed as an adjustment to new techniques developments. The second is how this changes the biologically equivalent dose per case, as related to an increased beam on time, delivered dose, etc., and how this potentially changes the radiobiological effects of GKRS in an individual patient.

Validation of PTV margin for Gamma Knife Icon frameless treatment using a PseudoPatient® Prime anthropomorphic phantom

The Gamma Knife Icon allows the treatment of brain tumors mask-based single-fraction or fractionated treatment schemes. In clinic, uniform axial expansion of 1 mm around the gross tumor volume (GTV) and a 1.5 mm expansion in the superior and inferior directions are used to generate the planning target volume (PTV). The purpose of the study was to validate this margin scheme with two clinical scenarios: (a) the patient's head remaining right below the high-definition motion management (HDMM) threshold, and (b) frequent treatment interruptions followed by plan adaptation induced by large pitch head motion. A remote-controlled head assembly was used to control the motion of a PseudoPatient® Prime head phantom; for dosimetric evaluations, an ionization chamber, EBT3 films, and polymer gels were used. These measurements were compared with those from the Gamma Knife plan. For the absolute dose measurements using an ionization chamber, the percentage differences for both targets were less than 3.0% for all scenarios, which was within the expected tolerance. For the film measurements, the two-dimensional (2D) gamma index with a 2%/2 mm criterion showed the passing rates of ≥87% in all scenarios except the scenario 1. The results of Gel measurements showed that GTV (D₁₀₀ ) was covered by the prescription dose and PTV (D₉₅ ) was well above the planned dose by up to 5.6% and the largest geometric PTV offset was 0.8 mm for all scenarios. In conclusion, the current margin scheme with HDMM setting is adequate for a typical patient's intrafractional motion.

Dosimetric evaluation of the Leksell GammaPlan™ Convolution dose calculation algorithm

The dosimetric accuracy of the Leksell GammaPlan Convolution calculation algorithm was evaluated through comparison with corresponding Monte Carlo (MC) dosimetric results. MC simulations were based on generated sector phase space files for the 4 mm, 8 mm and 16 mm collimator sizes, using a previous comprehensive Gamma Knife Perfexion^™ source model and validated using film dosimetry. Test cases were designed for the evaluation of the Convolution algorithm involving irradiation of homogeneous and inhomogeneous phantom geometries mimicking clinical cases, with radiation fields created using one sector (single sector), all sectors with the same (single shot) or different (composite shot) collimator sizes. Dose calculations using the Convolution algorithm were found to be in excellent agreement (gamma pass rate greater than 98%, applying 1%/1 mm local dose difference and distance agreement criteria), with corresponding MC calculations, indicating the accuracy of the Convolution algorithm in homogeneous and heterogeneous model geometries. While of minor clinical importance, large deviations were observed for the voxels laying inside air media. The calculated beam on times using the Convolution algorithm were found to increase (up to 7%) relative to the TMR 10 algorithm currently used in clinical practice, especially in a test case mimicking a brain metastasis close to the skull, in excellent agreement with corresponding MC calculations.

Considering inhomogeneities in Gamma Knife treatment planning: Factors affecting the loss of prescription dose coverage†

OBJECT

To compare the consistency of the agreement between the Convolution and TMR10 algorithms using a homogeneous phantom and to identify target characteristics that lead to large changes in target isodose coverage when the Convolution algorithm is used in GammaPlan as opposed to the TMR10 algorithm.

METHODS

The IROC phantom end-to-end test was performed and RTDose for both the TMR10 and Convolution algorithm were submitted for comparison to the measurement. Treatment plans for 16 patients and 26 different targets were retrospectively re-calculated with the Convolution algorithm when originally planned with the TMR10 algorithm. Multivariate regression was used to find statistically significant predictors of loss in target prescription isodose coverage.

RESULTS

Both algorithms agreed well with the IROC TLD measurement (within 1 %) and slightly better agreement was seen in the film analysis for the Convolution algorithm. After multivariate regression, small target volumes, < 1cm from air cavity, and minimum dose to target were potential predictors of large percentage loss of prescription isodose coverage (p = 0.049, 0.026, and 0.002, respectively).

CONCLUSION

Convolution and TMR10 appear to be equivalent in homogeneous situations. Some target characteristics have been identified that might be indications for use of the Convolution algorithm in clinical practice.

Factors Associated with Treatment Failure and Radiosurgery-Related Edema in WHO Grade 1 and 2 Meningioma Patients Receiving Gamma Knife Radiosurgery

BACKGROUND

Before the advent of radiosurgery, neurosurgical treatment of meningiomas typically involved gross total resection of the mass whenever surgery was deemed possible. Over the past 4 decades, though, Gamma Knife radiosurgery (GKRS) has proved to be an effective, minimally invasive means to control the growth of these tumors. However, the variables associated with treatment failure (regrowth or clinical progression) after GKRS and GKRS-related complications, such as cerebral edema, are less well understood.

METHODS

We retrospectively collected data between 2009 and 2018 for patients who underwent GKRS for meningiomas. After data collection, we performed univariate and multivariable modeling of the factors that predict treatment failure and cerebral edema after GKRS. Hazard ratios (HR) and P values were determined for these variables.

RESULTS

Fifty-two patients were included our analysis. The majority of patients were female (38/52,73%), and nearly all patients presented with a suspected or confirmed World Health Organization grade 1 meningioma (48/52, 92%). The median tumor volume was 3.49 cc (range, 0.22-20.11 cc). Evidence of meningioma progression after treatment developed in 5 patients (10%), with a median time to continued tumor growth of 5.9 months (range, 2.7-18.3 months). In multivariable analysis, patients in whom treatment failed were more likely to be male (HR = 8.42, P = 0.045) and to present with larger tumor volumes (HR = 1.27, P = 0.011). In addition, 5 patients (10%) experienced treatment-related cerebral edema. On univariate analysis, patients who experienced cerebral edema were more likely present with larger tumors (HR = 1.16, P = 0.028).

CONCLUSIONS

Increasing meningioma size and male gender predispose to meningioma progression after treatment with GKRS. Increasing tumor size also predicts the development of postradiosurgery cerebral edema.

Dosimetric comparison of TMR10 and convolution dose calculation algorithms in GammaPlan treatment planning system

Aims:

In this article, our goal is to compare the TMR10 and convolution dose calculation algorithm in GammaPlan used in stereotactic radiosurgery (SRS) treatments with Gamma Knife and to assess if the algorithms produce clinically significant differences.

Materials and methods:

Treatment plans were analysed from ten patients who have undergone Gamma Knife SRS treatments. Patient plans were retrospectively recalculated using Lesksell GammaPlan 10 treatment software utilising the TMR10 and convolution dose calculation algorithms in order to create a paired dataset for comparison. Evaluation was based on the dose volume histogram (parameters of minimum, mean, maximum and integral doses.

Results:

The ratios of average integral doses calculated by the convolution dose calculation algorithm to the average integral doses calculated by the TMR10 algorithm are 0·997 for the target (p=0·028), 1·048 (p=0·48) for the skull and 1·005 (p=0·68) for the brainstem.

Conclusions:

Although doses calculated with the convolution algorithm resulted in slightly higher mean integral doses for the brainstem and skull critical structures when compared to that of TMR10 doses, these results were not statistically or clinically significant. Thus we continue to use the TMR10 algorithm at our clinic.

Independent dose validation system for Gamma Knife radiosurgery, using a DICOM-RT interface and Geant4

Leksell GammaPlan was specifically designed for Gamma Knife (GK) radiosurgery planning, but it has limited accuracy for estimating the dose distribution in inhomogeneous areas, such as the embolization of arteriovenous malformations. We aimed to develop an independent patient dose validation system based on a patient-specific model, constructed using a DICOM-RT interface and the Geant4 toolkit. Leksell Gamma Knife Perfexion was designed in Geant4.10.00 and includes a DICOM-RT interface. Output factors for each collimator in a sector and dose distributions in a spherical water phantom calculated using a Monte Carlo (MC) algorithm were compared with the output factors calculated by the tissue maximum ratio (TMR) 10 algorithm and dose distributions measured using film, respectively. Studies using two types of water phantom and two patient simulation cases were evaluated by comparing the dose distributions calculated by the MC, the TMR and the convolution algorithms. The water phantom studies showed that if the beam size is small and the target is located in heterogeneous media, the dose difference could be up to 11%. In the two patient simulations, the TMR algorithm overestimated the dose by about 4% of the maximum dose if a complex and large bony structure was located on the beam path, whereas the convolution algorithm showed similar results to those of the MC algorithm. This study demonstrated that the in-house system could accurately verify the patient dose based on full MC simulation and so would be useful for patient cases where the dose differences are suspected.

A comparison of the convolution and TMR10 treatment planning algorithms for Gamma Knife® radiosurgery

AIMS

To compare the accuracies of the convolution and TMR10 Gamma Knife treatment planning algorithms, and assess the impact upon clinical practice of implementing convolution-based treatment planning.

METHODS

Doses calculated by both algorithms were compared against ionisation chamber measurements in homogeneous and heterogeneous phantoms. Relative dose distributions calculated by both algorithms were compared against film-derived 2D isodose plots in a heterogeneous phantom, with distance-to-agreement (DTA) measured at the 80%, 50% and 20% isodose levels. A retrospective planning study compared 19 clinically acceptable metastasis convolution plans against TMR10 plans with matched shot times, allowing novel comparison of true dosimetric parameters rather than total beam-on-time. Gamma analysis and dose-difference analysis were performed on each pair of dose distributions.

RESULTS

Both algorithms matched point dose measurement within ±1.1% in homogeneous conditions. Convolution provided superior point-dose accuracy in the heterogeneous phantom (-1.1% v 4.0%), with no discernible differences in relative dose distribution accuracy. In our study convolution-calculated plans yielded D_99% 6.4% (95% CI:5.5%-7.3%,p<0.001) less than shot matched TMR10 plans. For gamma passing criteria 1%/1mm, 16% of targets had passing rates >95%. The range of dose differences in the targets was 0.2-4.6Gy.

CONCLUSIONS

Convolution provides superior accuracy versus TMR10 in heterogeneous conditions. Implementing convolution would result in increased target doses therefore its implementation may require a revaluation of prescription doses.