Improved dose homogeneity using electronic compensation technique for total body irradiation

Abstract In total body irradiation (TBI) utilizing large parallel‐opposed fields, the manual placement of lead compensators has conventionally been used to compensate for the varying thickness throughout the body. The goal of this study is to pursue utilizing the modern electronic compensation (E‐comp) technique to more accurately deliver dose to TBI patients. Bilateral parallel‐opposed TBI treatment plans were created using E‐comp for 15 patients for whom CT data had been previously acquired. A desirable fluence pattern was manually painted within each field to yield a uniform dose distribution. The conventional compensation technique was simulated within the treatment planning system (TPS) using a field‐in‐field (FIF) method. This allows for a meaningful evaluation of the E‐comp technique in comparison to the conventional method. Dose–volume histograms (DVH) were computed for all treatment plans. The mean total body dose using E‐comp deviates from the prescribed dose (4 Gy) by an average of 2.4%. The mean total body dose using the conventional compensation deviates from the prescribed dose by an average of 4.5%. In all cases, the mean body dose calculated using E‐comp technique deviates less than 10% from that of conventional compensation. The average reduction in maximum dose using E‐comp compared to that of the conventional method was 30.3% ± 6.6% (standard deviation). In all cases, the s‐index for the E‐comp technique was lower (10.5% ± 0.7%) than that of the conventional method (15.8% ± 4.4%), indicating a more homogenous dose distribution. In conclusion, a large reduction in maximum body dose can be seen using the proposed E‐comp technique while still producing a mean body dose that accurately complies with the prescription dose. Dose homogeneity was quantified using s‐index which demonstrated a reduction in hotspots with E‐comp technique. Electronic compensation technique is capable of more accurately delivering a total body dose compared to conventional methods.


| INTRODUCTION
In total body irradiation (TBI) the goal is to deliver a uniform dose to the patient's whole body. 1 This serves to deplete the bone marrow and suppress the immune system of the patient as well as eliminate malignant cells. 2 The uniform delivery of a total body dose encounters challenges that are not seen in more standard external beam radiation therapy procedures. In TBI the fields must exceed the scattering volume (patient's whole body) in all directions and the human body is irregularly shaped which makes achieving a uniform dose difficult. There have been a number of clinical setups and methods of compensation proposed to overcome these challenges. 3 planning systems. This advancement has allowed for the capability to create a patient specific treatment plans which account for variables such as tumor size, shape, and tissue density. In contrast, the conventional treatment planning, setup, and delivery of TBI has yet to incorporate the advances of the field into a standard protocol. No procedural recommendations have been made since AAPM Report No. 17, 9 almost 30 yr ago.
Electronic compensation is a forward planned technique that utilizes the dynamic MLC to deliver the beam and replaces the use of a mechanical compensator. 10 It employs the use of the dynamic multileaf collimator (dMLC) to achieve a homogeneous dose distribution when irradiating irregular surfaces. Within a standard application of electronic compensation, the "Irregular shape compensator" module within the Varian ECLIPSE treatment planning platform (version 10 with AAA 10.0.28; Varian Medical Systems, Palo Alto, CA, USA) is used to calculate the optimal field fluence based on CT data. This function calculates the fluence such that a homogeneous dose is delivered to a user specified depth. Once the optimal fluence is calculated, it is converted into a deliverable fluence and then translated into MLC leaf motion by the leaf motion calculator (LMC). 10 We present here a comparison of our novel electronic compensation method to simulated conventional compensation for the delivery of a total body dose. By combining multi-leaf collimator with a three-dimensional treatment planning system, our work seeks to produce a dynamic photon beam that can generate an optimized fluence pattern that results in a more uniform TBI dose distribution.
This research aims to demonstrate that modifications to typical electronic compensation implementations can allow the technology to be adapted to the constraints of TBI and improve upon dosimetric accuracy of treatment delivery in comparison to standard TBI compensation methods.

2.A | General procedure
Fifteen previously treated patients were selected for whom CT data, from head to mid-calf, had been previously acquired. These patients comparison. This yields a more accurate representation of the dose distributions within a heterogeneous body when the MUs are calculated without considering a variation in tissue density. Each field was assigned a specific weight which was calculated based on the number of layers of lead used to compensate for that particular bilateral separation of the specific patient at the time of conventional delivery. With the knowledge of the linear attenuation coefficient of the lead compensators, the transmission was calculated for each differential thickness of lead. For patient 13 the lead compensators had an attenuation coefficient of 0.05406 and the transmission was calculated for five different thicknesses of lead (Table 1). Once the desired transmission was calculated for each field, it was used to determine the weight which would be assigned in the TPS. A field weight of 1.0 signifies that the field is open throughout the entire treatment and 0.0 would indicate that it is blocked throughout the treatment. In this study, the fields are executed in succession from the largest to the smallest with each of the smaller fields encompassed by the previous larger field. Beginning with the smallest field, which has a transmission of 1.0, the weight is determined by subtracting the transmission of the next largest field.

2.C | Electronic compensation for TBI
Because of the extended distance at which TBI is delivered, and the necessary extension of the field beyond scattering volume in all directions, the current Eclipse TM TPS module cannot calculate the optimal fluence as it does for cases that have a more standard geometry. Therefore, a desirable fluence pattern was manually painted within each field for each patient (Fig. 3). To achieve an

2.D | Comparison of electronic vs conventional compensation
The conventional planning and delivery techniques of TBI were compared to the application of electronic compensation to TBI delivery.
Cumulative dose-volume histograms (DVH) were generated to qualitatively compare the two methods (Fig. 4). The mean total body dose and maximum dose within the body were recorded for all four plans for each case. The s-index was used to quantitatively evaluate the merit of each technique. The s-index is a dose-volume histogram (DVH)-based homogeneity index that is defined as the standard deviation of the normalized differential DVH (dDVH) curve. 11 The dDVH is a plot of the volume receiving a dose within a specified dose range. Although less commonly used than the integral DVH curve, the differential DVH curve can be useful in that it can provide information regarding the extent of dose variation within a structure.
The standard deviation of the dDVH curve quantifies the spread of the average giving an indication the inhomogeneity of the dose. A larger standard deviation of the dDVH curve correlates to a more sizeable spread and, therefore, a less homogeneous dose distribution. 11 3 | RESULTS AND DISCUSSION

3.A | Materials minimal variation in mean body
dose -TDC on vs TDC off With respect to mean body dose, both the conventional simulation and the electronic compensation technique, with and without heterogeneity corrections, were able to deliver the prescription dose (4 Gy) with good accuracy as seen in Fig. 5. In all cases, the mean body dose from plans using electronic compensation deviates less than 7.1% from the prescription dose.

3.B | Dosimetric comparison of electronic vs conventional compensation
There was an evaluation of the electronic compensation technique in comparison to the conventional simulation. In every plan that was generated in this study, the maximum body dose was reduced using the electronic compensation technique (Fig. 6). The average reduction in Dose (Gy) Ratio of Total Structure Volume F I G . 4. Cumulative DVH of mean body dose cumulative DVH of the dose within the body of patient 13. The line with squares represents the body dose for the plan that utilized the electronic compensation method. The line with triangles represents the body dose for the plan that simulates the conventional treatment method. maximum dose was 30.3% with a standard deviation of 6.6% ( Table 2).
The mean total body dose for the plans utilizing the electronic compensation technique remained within 10% of the mean body dose for the conventional simulation ( Table 3). As seen in Table 3, the mean total body dose using electronic compensation deviates from the prescribed dose (4 Gy) by an average of 2.4%. The mean total body dose in the plans simulating the conventional compensation deviates from the prescribed dose (4 Gy) by an average of about 4.5%.
An important result of this study is the reduction in maximum body dose that is seen with the use of the electronic compensation technique for bilateral TBI. The goal of TBI is uniform dose distribution that is as close to the prescription dose as possible. 1 Any part of the body that receives a significantly higher dose than the prescribed 4 Gy, is considered a hotspot. Therefore the reduction in the size and intensity of hotspots is a desired effect of a new delivery technique. For every patient investigated within this study, the maximum body dose was significantly reduced with the use of electronic compensation. In the plans that simulate conventional delivery technique, the "hotspots" are typically found in the portion of the abdomen most anterior to the umbilicus, as seen in Fig. 7. This is a direct result of two-dimensional compensation. The amount of radiation received by this segment of the patient is based solely on the bilateral separation at midplane of the umbilicus without taking into consideration the tapering of the separation of the abdomen anterior to the midplane. With the use of electronic compensation, the amount of radiation received by any part of the patient is based on desired fluence in that particular area. The reduction in the size of the abdominal hotspot when using electronic compensation can be seen qualitatively in Fig. 7. These results suggest that the dose distribution for the electronic compensation technique is more homogenous than that of the conventional simulation.

3.C | Quantitative S-index comparison of electronic vs conventional compensation
To quantify the dose homogeneity, the s-index was used for comparison between electronic and conventional compensation for TBI.
In all 15 cases that were analyzed, the s-index for the electronic compensation technique was lower, indicating a more homogenous distribution (Fig. 8). The average s-index for the conventional simulation was 15.8% with a standard deviation of 4.4%. The average s-index for the electronic compensation was 10.5% with a standard deviation of 0.7%. In Fig. 8, the solid lines represent the mean s-index for each technique and the dashed lines show the standard deviation for each set of data. There are two patients for whom the s-index of the conventional simulation is significantly higher than for any other patients. This indicates a poor homogeneity within the dose distribution for these plans. This is confirmed when we consider the DVH for each of these plans (Fig. 9).

3.D | Qualitative comparison of electronic vs conventional compensation
The electronic compensation technique would improve the accuracy with which we are able to deliver a prescribed total body dose.
The conventional method relies on the ability to match a shadow that is projected onto the patient, using the light field and lead compensation, to the corresponding section of the anatomy that is planned to receive that amount of compensation. This could prove to be particularly difficult in case of the neck/shoulder juncture. pensation. This study also indicates that the use of electronic compensation in the planning and delivery of TBI provides a mean total body dose that more accurately complies with the prescribed dose. In summary, this study illustrates that an electronic compensation technique is capable of more accurately delivering a total body dose compared to current conventional methods.

CONFLI CT OF INTEREST
The authors declare no conflict of interest.  F I G . 9. dDVH Patient 8 conventional dDVH (a) and the dDVH for the electronic compensation technique (b). Patient 11 conventional dDVH (c) and the dDVH for the electronic compensation technique (d). The x-axis represents Dose (Gy) and the y-axis represents dVolume/dDose (cm 3 /Gy).