S Dhanesar

The role of Cobalt-60 in modern radiation therapy: Dose delivery and image guidance.

The advances in modern radiation therapy with techniques such as intensity-modulated radiation therapy and image-guided radiation therapy (IMRT and IGRT) have been limited almost exclusively to linear accelerators. Investigations of modern Cobalt-60 (Co-60) radiation delivery in the context of IMRT and IGRT have been very sparse, and have been limited mainly to computer-modeling and treatment-planning exercises. In this paper, we report on the results of experiments using a tomotherapy benchtop apparatus attached to a conventional Co-60 unit. We show that conformal dose delivery is possible and also that Co-60 can be used as the radiation source in megavoltage computed tomography imaging. These results complement our modeling studies of Co-60 tomotherapy and provide a strong motivation for continuing development of modern Cobalt-60 treatment devices.

Practical and clinical considerations in Cobalt-60 tomotherapy

Cobalt-60 (Co-60) based radiation therapy continues to play a significant role in not only developing countries, where access to radiation therapy is extremely limited, but also in industrialized countries. Howver, technology has to be developed to accommodate modern techniques, including image guided and adaptive radiation therapy (IGART). In this paper we describe some of the practical and clinical considerations for Co-60 based tomotherapy by comparing Co-60 and 6 MV linac-based tomotherapy plans for a head and neck (HandN) cancer and a prostate cancer case. The tomotherapy IMRT plans were obtained by modeling a MIMiC binary multi-leaf collimator attached to a Theratron-780c Co-60 unit and a 6 MV linear accelerator (CL2100EX). The EGSnrc/BEAMnrc Monte Carlo (MC) code was used for the modeling of the treatment units with the MIMiC collimator and EGSnrc/DOSXYZnrc code was used for beamlet dose data. An in-house inverse treatment planning program was then used to generate optimized tomotherapy dose distributions for the H and N and prostate cases. The dose distributions, cumulative dose area histograms (DAHs) and dose difference maps were used to evaluate and compare Co-60 and 6 MV based tomotherapy plans. A quantitative analysis of the dose distributions and dose-volume histograms shows that both Co-60 and 6 MV plans achieve the plan objectives for the targets (CTV and nodes) and OARs (spinal cord in HandN case, and rectum in prostate case).

Cobalt-60 tomotherapy: Clinical treatment planning and phantom dose delivery studies

PURPOSE Investigations have shown that a Cobalt-60 (Co-60) radioactive source has the potential to play a role in intensity modulated radiation therapy (IMRT). In this paper, Co-60 tomotherapys conformal dose delivery potential is evaluated by delivering conformal dose plans on a cylindrical homogeneous phantom containing clinical structures similar to those found in a typical head and neck (H&N) cancer. Also, the clinical potential of Co-60 tomotherapy is investigated by generating 2D clinical treatment plans for H&N and prostate anatomical regions. These plans are compared with the 6 MV based treatment plans for modalities such as linear accelerator-based tomotherapy and broad beam IMRT, and 15 MV based 3D conformal radiation therapy (3DCRT). METHODS For experimental validation studies, clinical and nonclinical conformal dose patterns were delivered on circular, homogeneous phantoms containing GafChromic film. For clinical planning study, dose calculations were performed with the EGSnrc Monte Carlo program, where a Theratronics 780C Co-60 unit and a 6 MV linear accelerator were modeled with a MIMiC binary multileaf collimator. An inhouse inverse treatment planning system was used to optimize tomotherapy plans using the same optimization parameters for both Co-60 and 6 MV beams. The IMRT and 3DCRT plans for the clinical cases were generated entirely in the Eclipse treatment planning system based on inhouse IMRT and 3DCRT site specific protocols. RESULTS The doses delivered to the homogeneous phantoms agreed with the calculations, indicating that it is possible to deliver highly conformal doses with the Co-60 unit. The dose distributions for Co-60 tomotherapy clinical plans for both clinical cases were similar to those obtained with 6 MV based tomotherapy and IMRT, and much more conformal compared to 3DCRT plans. The dose area histograms showed that the Co-60 plans achieve the dose objectives for the targets and organs at risk. CONCLUSIONS These results confirm that Co-60 tomotherapy is capable of providing state-of-the-art conformal dose delivery and could be used for the treatment of targets in both small and larger separation anatomical regions.

Aperture superposition dose model versus pencil beam superposition dose model for a finite size Cobalt-60 source for tomotherapy deliveries

PURPOSE The finite size pencil beam (FSPB) superposition method is a commonly used dose calculation method in intensity modulated radiation therapy (IMRT). The FSPB model assumes that dose for a broad intensity modulated beam can be calculated by superposition of dose from small, pencil-like beams. However, this model is limited to point-like radiation sources and is not valid for finite size sources, such as a Cobalt-60 (Co-60) source of 2 cm diameter. In this paper, the authors present results that show the limitation of this model and propose an alternative model, namely the aperture superposition (AS) model, to calculate photon dose for intensity modulated beams arising from finite size radiation sources. METHODS The AS model is based on adding beam apertures rather than pencil beams. Each aperture is defined as a series of adjacently opened leaves of a multileaf collimator with no closed leaves in between them. The apertures are calculated using the EGSnrc Monte Carlo program. The accuracy of the AS model was tested for dose calculations of fan beams, as encountered in tomotherapy treatment plans. The results were compared with the FSPB model and GafChromic film measurements. The measurements and simulations were performed for a clinical Theratronics T780C Co-60 unit with MIMiC binary multileaf collimator mounted on it. RESULTS The comparisons between the AS model and film measurements show agreement better than 1.5% in the high dose regions and 3.7% in the low dose regions. On the contrary, film measurement comparisons to the FSPB model show that the FSPB model underestimates the dose by up to 7% for small field sizes such as 2 × 2 cm(2) and 20% for larger field sizes such as 20 × 2 cm(2). CONCLUSIONS The results presented in this paper indicate that the AS model provides better accuracy than the FSPB model when calculating dose for fan beams from large radiation sources. The implementation of this model to the current treatment planning systems has the scope of advancing Co-60 based IMRT and tomotherapy.

SU‐EE‐A1‐06: A Comparative Study of Cobalt‐60 Based Tomotherapy versus 6 MV Linac‐Based Tomotherapy, IMRT, and 3DCRT for the Treatment Planning of Prostate and Head and Neck Cases

Purpose: Cobalt‐60 (Co‐60) based radiation therapy continues to play a significant role in a large number of countries due its simplicity and robustness. However, it has not been developed to accommodate modern techniques that provide intensity modulated radiation therapy(IMRT). In this paper we present the results of investigations of Co‐60 based tomotherapy. Particularly, we generate clinical plans for prostate and head and neck (H&N) anatomical regions and compare them with the plans obtained with 6MV based linac tomotherapy, standard 6MV IMRT, and 3D conformal radiation therapy (3DCRT) techniques. Method and Materials: The tomotherapy plans were obtained by modeling a MIMiC binary multileaf collimator attached to a Theratron‐780C Co‐60 unit and a 6MV linear accelerator. The EGSnrc/BEAMnrc Monte Carlo code was used to model the treatment units with the MIMiC collimator while EGSnrc/DOSXYZnrc code was used for calculating dose on prostate and H&N CT datasets. All heterogeneities and patient contours were considered. An in‐house inverse treatment planning program was then used to optimize all 2D tomotherapy plans. The IMRT and 3DCRT plans were generated in Eclipse treatment planning system based on our in‐house IMRT and 3DCRT clinical protocols for prostate and H&N treatment.Results: A quantitative analysis of the dose distributions and dose area histograms (DAHs) showed that the Co‐60 plans achieve the dose objectives for the targets and OARs. The dose distributions and DAHs for Co‐60 tomotherapy plans for both cases are very similar to those obtained with 6MV based tomotherapy and IMRT, and are much more conformal compared to 3DCRT plans. Conclusion: Our investigations confirm that Co‐60 tomotherapy is indeed capable of providing state‐of‐the‐art conformal dose delivery and could be used for the treatment of targets in both small and larger separation anatomical regions.

TH‐C‐AUD‐08: Comparison of Tomotherapy Dose Distributions for 6MV X‐Rays and Different Cobalt‐60 Source Designs Using Monte Carlo Methods

Purpose: To investigate intensity modulated dose distributions for Co‐60 based tomotherapy with cylindrical and rectangular shaped source geometries for a typical head and neck case using Monte Carlo(MC) methods. Method and Materials: EGSnrc/BEAMnrc MC code has been used to model three Co‐60 tomotherapy units and a 6 MV (Varian 2100EX) unit. Two of the Co‐60 units, consisting of modified collimator systems with customized binary multi‐leaf collimator (BMLC) were modeled for 3 and 5 cm long rectangular sources and SADs of 70 and 80 cm, respectively. The other units were a conventional Co‐60 unit (T780c) with 2 cm diameter cylindrical source (MDS Nordion, Canada) and a 6 MV Linac with an addon MIMiC BMLC (NOMOS, USA). All the Co‐60 sources had identical active volumes. EGSnrc/DOSXYZnrc MC code was used to calculate the intensity modulated beamlets and fan beam dose profiles in a water phantom. The intensity modulated energy fluence profiles from a 6 MV Linac and two Co‐60 units using the same beam segments will be compared. Tomotherapy treatment plans for a typical H&N case were calculated. All plans were optimized for PTV, two neck nodes and the spinal cord using the same optimization parameters. The treatment planning was provided by an in‐house inverse planning system based on a multi‐objective gradient‐search approach using MC calculated dose data. Results: Comparisons of optimized dose distributions, dose difference maps and dose area histogram will be presented. Despite significant differences in fluence profiles for Co‐60 and 6 MV beams, optimized dose distribution for all plans met the dose‐volume and tolerance criteria. However, the integral doses are potentially higher for Co‐60 plans particularly with longer source size and shorter SAD.Conclusions: Co‐60 based tomotherapy with appropriate source and BMLC design is dosimetrically viable. Research supported (in‐kind) by MDS Nordion, Canada.

SU‐E‐T‐558: Photon Fluence Model for Distributed Radiation Sources Using the Convolution Method

PURPOSE Sophisticated dose calculation algorithms based on principles of convolution theory are commonly used in commercial treatment planning systems for calculating photon dose from point-like radiation sources. However, not much has been studied regarding the modeling of distributed sources such as a Cobalt-60 (Co-60) source of 2 cm diameter. In this work, we present a convolution based photon fluence model that can be used for dose calculations for any finite size source. METHODS The photon fluence model proposed in this work is based on two key functions: the source distribution function and the aperture function. The source distribution function, which models primary and scatter radiation independently, was determined using a one-time Monte Carlo (MC) simulation. Since the focus of this work is on Co-60 delivery, the MC modeling was based on a Theratronics T780 Co-60 unit (Best Theratronics). The validity of the MC model was verified with the ion chamber measurements made in a water phantom. The aperture function is a rectangular window function representing the field size. The fluence model was evaluated by comparing the calculated photon fluence to the MC simulated fluence. RESULTS The comparisons between the photon fluence calculated using the proposed model and the MC simulation fluences show good agreement, better than 2% in the in-field region of all large and small fields. In the tail and sharp dose gradient regions, this agreement is better than 5% for the large field sizes and 1.5% for the small field sizes. The primary and scatter fluence output factors are within 2% of those obtained with the MC simulations. CONCLUSION The results of our investigations indicate that the proposed photon fluence model can accurately determine fluence for large and small radiation beams from finite size sources. This model has the potential of playing an integral role in Co-60 based IMRT treatment planning.

SU-E-T-745: Convolution-Superposition Model for Photon Dose Calculations of Finite Size Cobalt-60 Radiation Source

Purpose: The collapsed cone convolution (CCC) superposition method is commonly used to calculate dose in intensity modulated radiation therapy(IMRT), particularly for beams originating from a point source of radiation. Here we propose and present a validation of a modified version of this method that can be used for dose calculations of intensity modulated beams from finite size radiationsources such as Cobalt‐60 (Co‐60) source. Methods: The CCC method computes 3D dose by convolving the TERMA (total energy released in a medium per unit mass), which is dependent on energy fluence, with a photondose kernel. In our method, TERMA is calculated using an approach that takes into account the effective source diameter. Specifically, the energy fluence depends on a sourcedistribution function, which is based on calculating the effective source diameter for a particular field size. The calculations of the modified convolution model were compared with the GafChromic film measurements and EGSnrc Monte Carlo (MC) calculations. The studies were done on a clinical Theratronics 780C Co‐60 unit with a 2cm diameter cylindrical source. Results: The energy fluence for various field sizes ranging from 1×1cm2 to 30×30cm2 was calculated and compared with the MC simulations. The results showed agreement better than 1.4% for fields centered on the central‐axis and 3% for those centered off‐axis. The dose was calculated by convolving energy fluence with the MC based pre‐calculated dose kernels. The dose comparisons to film measurements showed agreement to 2% in high dose regions and 3.8% in low dose penumbral regions. Conclusions: The results of this study show that the modified convolution‐superposition method can provide an acceptable accuracy when calculating dose for finite size sources. The implementation of this model to the current treatment planning systems can be useful for treatment planning of Co‐60 based IMRT and tomotherapy. Ontario government funding through Ontario Consortium for Adaptive Interventions in RadiationOncology (OCAIRO), which has an industrial component of matched support from Best Theratronics (Kanata, ON).

Sci—Thur PM: YIS — 04: Aperture Superposition Algorithm for Photon Dose Calculations of Finite Size Cobalt‐60 Radiation Source for Tomotherapy Dose Deliveries

The finite size pencil beam (FSPB) superposition method is commonly used to calculate dose for intensity modulated beams in IMRT. The FSPB model assumes that broad beam dose from a radiation originating from a point source can be calculated by a superposition of dose from pencil beams. For finite size sources, such as a Cobalt‐60 (Co‐60) source of 2 cm diameter, this method is no longer valid. In this paper we propose an aperture superposition (AS) dose calculation method that can be used for dose calculations of intensity modulated beams from finite size radiation sources. The model is applied to fan beams, as encountered in tomotherapy, and results are compared to the FSPB model and the film measurements. The comparisons between the AS model and film measurements show agreement to 1.5% in the high dose regions and 3.7% in the low dose regions. On the other hand, film measurement comparisons to the FSPB model show that the FSPB model underestimates the dose by up to 7% for small field sizes such as 2×2cm2 and 20% for larger field sizes such as 20×2 cm2. In conclusion, the AS model provides a better accuracy than the FSPB model when calculating dose for fan beams from large radiation sources. Research is underway to extend the application of this model to broad IMRT beams obtained from non‐binary multi‐leaf collimators. The implementation of this model to the current treatment planning systems can be useful for treatment planning of Co‐60 based IMRT and tomotherapy.

Poster — Wed Eve—34: Design of a Primary Collimator for Cone Beam CT Imaging

Intensity Modulated Radiation Therapy(IMRT) is becoming the standard of care for a number of cancer sites. Our research work has focused on developing Co‐60 based IMRT as an alternative for those areas in the world with limited infrastructure for supporting LINAC based systems. We have, to date, considered Co‐60 IMRT based on the tomotherapy approach because the full rotational delivery properties of tomotherapy helps to overcome the problems associated with the limited penetration of Co‐60 beams. However, with the recent introduction of arc based broad beam IMRT technology such as RapidArc (Varian Medical Systems, Palo Alto, CA), there is a potential for Co‐60 to be used in this approach. One potential advantage of broad beam versus fan beam Co‐60 IMRT delivery would be a reduction in the beam‐on time; a feature particularly important for a source that decays. Cone beam CT(CBCT)imaging is widely used for image guidance of broad‐beam IMRT and could likely play a role in a Co‐60 based broad beam IMRT system. The simplest and least complex design for such a system would consist of a single Co‐60 source for therapy and CBCTimaging. In this report we describe and analyze a new multileaf variable aperture primary collimator that permits a single Co‐60 source to be “switched” from therapy to a lower dose rate imaging mode. Monte carlo simulations of this design show that with a combination of altered aperture shape and an attenuator, an acceptable imagingdose rate without loss of spatial resolution is achieved.