|Year : 2013 | Volume
| Issue : 4 | Page : 618-623
Influence of the modulation factor on the treatment plan quality and execution time in Tomotherapy in head and neck cancer: In-phantom study
Adam Ryczkowski1, Tomasz Piotrowski2
1 Department of Medical Physics, Greater Poland Cancer Centre, Poznan, Poland
2 Department of Medical Physics, Greater Poland Cancer Centre; Department of Electroradiology, University of Medical Sciences, Poznan, Poland
|Date of Web Publication||11-Feb-2014|
Department of Medical Physics, Greater Poland Cancer Centre, Garbary 15 Street, 61-866 Poznan
Source of Support: None, Conflict of Interest: None
Purpose: The overall aim was to conduct an analytical study of the impact of the modulation factor (MF) on the quality of the head and neck treatment plans and their execution time on Tomotherapy.
Materials and Methods: In-phantom (RANDO® Alderson) planning study of the head and neck cancer was performed. Thirteen different plans in terms of MF were prepared. Other optimization parameters were the same for all plans.
Results: Analysis of treatment plans in terms of quality shows that MF < 1.4 does not provide an accepted dose distribution (physician decision). Statistically significant differences were observed for plans with an MF < 1.6. No differences were obtained for plans with MF from 6.0 to 1.8. Decreasing of MF leads to a shorter time of irradiation. The maximum rotational speed has been reached for an MF = 3.0. Further reducing this however produces no decrease in the time of irradiation. The actual and planned values of the MF were compared. The optimal range of MF for head and neck was determined as 3.0 > MF > 1.8. The lower limit increases to 2.4 when hard reduction of the dose in critical organs is required.
Conclusions: It was showed that the final MF value is less than the value calculated after each loop of optimization. The computer system reduces MF by shortening the longest time and increasing the average time of leaves opening. Increase in the average time is obtained by eliminating the use of leafs with the shortest times of opening, thereby reducing the dose in critical organs that are outside the direct irradiation area.
Keywords: Modulation factor, optimization, Tomotherapy, treatment planning
|How to cite this article:|
Ryczkowski A, Piotrowski T. Influence of the modulation factor on the treatment plan quality and execution time in Tomotherapy in head and neck cancer: In-phantom study. J Can Res Ther 2013;9:618-23
|How to cite this URL:|
Ryczkowski A, Piotrowski T. Influence of the modulation factor on the treatment plan quality and execution time in Tomotherapy in head and neck cancer: In-phantom study. J Can Res Ther [serial online] 2013 [cited 2019 Dec 12];9:618-23. Available from: http://www.cancerjournal.net/text.asp?2013/9/4/618/126458
| > Introduction|| |
A Tomotherapy machine combines two devices: Megavoltage therapeutic accelerator and computed tomography (CT) scanner. ,, To irradiate the patient from all possible directions, a 6 MV photon fan beam is used. It is modulated by a 64-leaf collimator. ,, The rotary motion of the gantry combined with the advancing motion of the table allows the radiation source to move around the patient along a spiral track.  On the opposite side of the gantry, there is a matrix of detectors, which enables the collection of CT images. ,
An "inverse planning" system is used for treatment planning. ,,,, At the beginning, some parameters, such as field width, pitch, modulation factor (MF), importance, and penalty for all structures, are determined. 
The user can choose among three field widths - 1.0, 2.5, and 5.0 cm - depending on the size of target.  This value specifies the size of the field in the machine isocenter along the longer axis of the patient.
The pitch parameter specifies by what part of the field width will the table move during one full rotation of the gantry. To obtain a homogeneous dose distribution, outside the isocenter, this value should meet the condition of 0.86/n, where n is an integer. 
MF is defined as:
MF = max open time/average open time (1)
But only the times greater than zero are averaged. For value of one, opening time for all the leaves is the same. 
Importance and penalty parameters are defined for each structure separately. They indicate the weight of individual structures in the process of optimization to the planning system. These parameters and MF may be modified during the optimization process of the treatment plan. 
Algorithms for selection parameters as field width and pitch are clearly defined. The problem may lie in determining the appropriate value of MF. Reducing its value results in shortening the duration of treatment; however, this may adversely affect the dose distribution. The literature reports on the use of MF for the head and neck area ranging from 1.8 up to 3.0. Kinhikar et al.  used the values of 2.0 and 2.8; Sheng's  article reports a value of 2.5; and Fiorino et al.  used an MF value of 3.0. It is worth noting that all the authors used the same values of the other two basic parameters. The width of the field in all cases was 2.5 cm and the pitch was 0.3. Moldovan et al. compared plans with varying field widths and pitches; however, they always used one value of MF.  However, none of the articles present the impact of MF on the quality of the treatment plans and their execution time.
The overall objective of this study was to analyze the impact of MF on obtained treatment plans and their execution time on the Tomotherapy machine. Specific objectives include preparing treatment plans for the head and neck area diversified in terms of value of MF and analyzing them in terms of quality and execution time of treatment with the Tomotherapy machine.
| > Materials and Methods|| |
To prepare the treatment plan, CT scans of anthropomorphic phantom Alderson RANDO® were used.  The head and neck area was scanned. Distance between the sections in the longitudinal axis of phantom was 3 mm.
Critical organs such as the brain, brainstem, spinal cord, eyes, lens, visual nerves, visual intersection (chiasm), parotid (right, left), mandible, oral cavity, larynx, and lung peaks were all included in the scans. A target corresponding to a nasopharyngeal cancer was included, too [Figure 1].
|Figure 1: Illustration of the main critical organs (parotid glands and spinal cord) and target volume on CT scans used in the study|
Click here to view
Contours of the structures were prepared in the Eclipse treatment planning system (Varian Medical Systems Inc., Palo Alto, CA, USA). This system is used on an everyday basis by radiotherapists in clinical practice.
Thus, the prepared tomography scans with contours of organs were sent to the TomoTherapy HiArt System (Accuray Inc., Sunnyvale, CA, USA), version 3.5.
Based on imported CT scans, a treatment plan was prepared. The main goal of the plan was to receive the dose of 50 Gy on 95% volume of the target.
The initial value of MF was 6.0. Other basic parameters were given values typical for treatment plans used for the head and neck area. Field width was 2.5 cm, the value of pitch was 0.215, and a grid was used to calculate the highest possible resolution (0.195 × 0.195 cm, fine option).
After calculating the beamlets, optimization of the treatment plan was begun. The primary aim was to reduce dose in the following organs: Spinal cord, parotid glands (right, left), mandible, larynx, and oral cavity. Other critical organs were outside the irradiated field, and therefore any received dose came from only scattered radiation. The final plan criteria are included in [Table 1].
To obtain the optimal dose distribution during the treatment plan optimization, the importance and penalty parameters were adjusted for target and critical organs.
These changes were made in response to the shape of the dose-volume histogram. The final values are shown in [Table 2]. Later, these parameters were not modified.
|Table 2: Final values of optimization parameters for target and critical organs|
Click here to view
Obtaining an acceptable treatment plan required 900 optimization loops made by the computer system. Each loop takes about 8-10 s, which, together with the time needed to change the parameters, makes the optimization time approximately 3.5 h. Then the dose was divided into fractions (25 fractions at 2 Gy each). After the final dose distribution was calculated by the system (including fractional dose), a treatment plan was archived.
The next step was to reduce the MF. Its value was reduced progressively until a minimum value (equal to 1.0) was reached.
The aim of reducing MF is to shorten the execution time of treatment. In practice, a person planning the treatment decreases the value of MF as long as it does not affect the quality of the obtained dose distribution.
The value of MF was changed as follows:
The range of 6.0-3.0 in step at 1.0.
The range of 2.6-1.0 in step at 0.2.
After each change of MF, the computer system performed 100 optimization loops. It was essential that the algorithm adjust the actual rate of MF to the new requirements, and hence change the weight of each beamlet to obtain a dose distribution consistent with the assumptions.
Finally, 13 plans with the same set of Importance and Penalty parameters were obtained, differing in the value of MF. Each plan was archived, the final dose-volume histograms were exported, and the report was printed. After gathering all the data, the effect of MF on the quality and duration of the treatment plan was analyzed.
For comparison, the average normalized dose in each treatment plan was t-tested for related variables. The distribution of relative average values of individual plans did not differ significantly from a normal distribution. Statistical testing was performed using StatSoft Statistica Version 8 (StatSoft Inc., Tulsa, OK, USA).
| > Results|| |
Initially, the change in the value of MF (in the range of 6.0-3.0) had no effect on the shape of the dose-volume histogram. After several iterations, the dose distribution was the same as the one before the change of the parameter.
In the range of 2.6-2.2, the change in MF initially caused large, readily visible differences in the obtained dose-volume histogram. However, 100 iterations performed by the computing system resulted in the dose distribution returning to its original form.
For MF below 2.0, the obtained plans did not return exactly to its original form.
After the final calculation of dose distribution, the dose-volume histogram from each plan was exported. Thus, it could be further analyzed in Microsoft Excel 2007.
[Figure 2] shows eight dose-volume histograms for the treatment plans for the selected MFs. They allow one to observe changes in the dose distribution in the target and in the critical organs with a change in the value of MF. For greater clarity, the curves for structures outside the irradiated area were omitted.
The descriptive statistics and dose-volume histograms obtained from the reports were used to analyze the impact of a fixed value of MF on the quality of the treatment plan.
|Figure 2: Dose-volume histograms for six critical organs and target, modulation factor 6.0, 3.0, 2.4, 2.0, 1.8, 1.6, 1.2, and 1.0|
Click here to view
The primary way to assess the quality of the treatment plan was to ensure whether they meet the initial criteria [Table 1]. According to them, the plans with an MF value below 1.4 would not be accepted (the limit of the average absorbed dose for the left parotid gland is exceeded, the remaining structures are close to the border of acceptance).
The initial reduction of the MF value had no noticeable effect on the obtained dose distribution. Therefore, statistical testing was carried out to determine under what values of MF treatment plans differ significantly.
For this purpose, the average absorbed dose was used. In [Figure 3], there are summarized graphs showing the absolute average absorbed dose (Dave ) for the target and six critical organs in relation to MF.
|Figure 3: Overview of the average absorbed dose for target and selected critical organs in relation to modulation factor|
Click here to view
For statistical analysis in Statistica 8, the relative values of the average absorbed dose were used. t-Tests for related variables were used, because the distribution of relative values did not differ significantly from normal distribution. The results of the test were analyzed at the significance level α = 0.05.
[Table 3] shows the results of the comparison of treatment plans for MFs of 3.0-1.0. For further transparency, the results of comparisons for the plans with MF greater than 3.0 were omitted (as they did not show statistically significant differences). The statistically significant results at the level α = 0.05 are marked in bold.
|Table 3: The results of t-test for related samples, P value, statistically significant (P value <0.05) marked in bold|
Click here to view
In addition to basic descriptive statistics, information about the duration of the irradiation process and the gantry rotation time were obtained from the report.
After the final calculation of the opening time for all leaves of the collimator in all projections, the computer system can calculate the real value of MF. This information is also included in this study.
Initial MF is only an upper bound for the optimization algorithm. In the final calculation of opening time of the leaves, the shortest beamlets that the machine is not able to perform are rejected. Then, the final MF, with the rate usually lower than the initial MF, is calculated.
[Table 4] summarizes the four values for the thirteen treatment plans: MF determined during the planning, actual MF calculated by the computer system, duration of the irradiation process (T), and gantry rotation time (Tg ).
|Table 4: Comparison of the planed value of the modulation factor with actual value, treatment time (T), and time of gantry rotation (Tg)|
Click here to view
[Figure 4] shows a graph representing the relation between the shortening of irradiation time and the decrease in the value of MF.
|Figure 4: Graph showing the relation between treatment time and planed modulation factor|
Click here to view
| > Discussion|| |
To achieve set objectives, thirteen treatment plans, which differed only in MF value, were created. Other parameters, such as field width, pitch, and the importance and penalty values for all structures, remained unchanged. This allowed one to analyze the effects of MF on the quality of the plan and the time of its implementation.
For MF values below 1.4, the plans did not meet the established criteria [Table 1]. Average dose for the left parotid gland (27.52 Gy) was higher than the limit value (26 Gy). For other structures, the values were close to crossing the border of acceptance.
For treatment plans with MFs of 1.2 and 1.0, dose distribution for the main target volume was also significantly worse. It became heterogeneous, as can be seen in the dose-volume histogram [Figure 2], the curve is not stepped] and in descriptive statistics (the difference between the maximum and minimum doses increases, standard deviation exceeds 1 Gy).
Statistical testing was conducted to determine under which value of MF the obtained plans differed significantly from the original treatment plan (with an MF = 6.0). The results of statistical testing [Table 3] revealed a clear difference. Treatment plans with MF in the range of 6.0-1.8 did not show a statistically significant difference (p value ≥ 0.05); the plans showed a significant difference when MF was set to 1.6 and less.
Based on the results of statistical tests, the value of 1.8 can be regarded as a lower limit of the optimum MF for the head and neck area. Analysis of the chart showing the average dose for the analyzed organs as a function of MF [Figure 3] indicates that the curves for the parotid gland and larynx begin to deviate from the horizontal line with the MF below 2.4.
The main reason for limiting the value of MF is to decrease the time required to perform a plan on the therapeutic machine. As expected, the duration of treatment for the developed plans decreased together with the decreasing value of MF [Table 4]. As can be seen in [Figure 4], this correlation of MF of 6.0-3.0 is approximately linear. However, for treatment plans with MF value below 3.0, the time of irradiation is not shortened. This is related to the reaching of the maximum possible speed of the machine gantry - its rotation time (Tg ) may not be less than 15 s. The value of 3.0 can be regarded as the upper limit of the optimal rate of modulation. For the latest version of the treatment planning system (v4.0), the minimum gantry rotation period is 12 s. This allows the execution time of treatment to be shortened to an MF of about 2.5.
After connecting the lower limit (resulting from an analysis of the developed treatment plans in terms of quality) with the upper limit (resulting from the analysis of the time of treatment plan execution), the optimum MF value for the head and neck area can be determined. The obtained data show that it is within the range of 3.0-1.8 (or 2.4 if it is necessary to limit the maximum dose in critical organs). It should be noted that the results were affected by parameters such as field width, pitch, and fractional dose, which in this case were 2.5 cm, 0.215, and 2 Gy, respectively. These are standard values used in treatment plans for the analyzed area.
The comparison of the actual MF, calculated by the system after the final dose distribution calculation, with the value entered by the user showed that the final MF is significantly lower. This is related to the rejection of leaves with a very short opening time by the computer algorithm. This increases the average time of leaf opening, which results in accordance with formula (1), reducing the MF.
| > Conclusion|| |
Analysis of the quality of the obtained treatment plans allows one to specify the lower limit of the optimum MF for the head and neck area at the level of 1.8. If it is necessary to reduce the dose in critical organs because of clinical reasons, the value should not be lower than 2.4.
Analysis of individual treatment plan time execution allows one to specify the upper limit of the optimal MF factor at the level of 3.0.
The comparison of actual MF values with the restriction introduced by the user showed that final value of MF is significantly lower.
| > References|| |
|1.||Mackie TR, Holmes T, Swerdloff S, Reckwerdt P, Deasy JO, Yang J, et al. Tomotherapy: A new concept for the delivery of dynamic conformal radiotherapy. Med Phys 1993;20:1709-19. |
|2.||Fenwick JD, Tomé WA, Soisson ET, Mehta MP, Rock Mackie T. Tomotherapy and other innovative IMRT delivery systems. Semin Radiat Oncol 2006;16:199-208. |
|3.||Murthy V, Master Z, Gupta T, Ghosh-Laskar S, Budrukkar A, Phurailatpam R, et al. Helical tomotherapy for head and neck squamous cell carcinoma: Dosimetric comparison with linear accelerator-based step-and-shoot IMRT. J Cancer Res Ther 2010;6:194-8. |
|4.||Yang JN, Mackie TR, Reckwerdt P, Deasy JO, Thomadsen BR. An investigation of tomotherapy beam delivery. Med Phys 1997;24:425-36. |
|5.||Budgell G. Intensity modulated radiotherapy (IMRT): An introduction. Radiography 2002;8:241-9. |
|6.||Malicki J. The importance of accurate treatment planning, delivery, and dose verification. Rep Pract Oncol Radiother 2012;17:63-5. |
|7.||Ivanova T, Bliznakova K, Malatara G, Kardamakis D, Kolitsi Z, Pallikarakis N. Simulation studies on the effect of absorbers on dose distribution in rotational radiotherapy. Phys Med 2009;25:172-80. |
|8.||Forrest LJ, Mackie TR, Ruchala K, Turek M, Kapatoes J, Jaradat H, et al. The utility of megavoltage computed tomography images from a helical tomotherapy system for setup verification purposes. Int J Radiat Oncol Biol Phys 2004;60:1639-44. |
|9.||Shah AP, Langen KM, Ruchala KJ, Cox A, Kupelian PA, Meeks SL. Patient dose from megavoltage computed tomography imaging. Int J Radiat Oncol Biol Phys 2008;70:1579-87. |
|10.||Chui CS, Spirou SV. Inverse planning algorithms for external beam radiation therapy. Med Dosim 2001;26:189-97. |
|11.||Holmes TW, Mackie TR, Reckwerdt P. An iterative filtered backprojection inverse treatment planning algorithm for tomotherapy. Int J Radiat Oncol Biol Phys 1995;32:1215-25. |
|12.||Kazmierska J, Malicki J. Application of the Naïve Bayesian Classifier to optimize treatment decisions. Radiother Oncol 2008;86:211-6. |
|13.||Kumar SAS, Vivekanandan N, Sriram P. A study on conventional IMRT and RapidArc treatment planning techniques for head and neck cancers. Rep Pract Oncol Radiother 2012;17:168-75. |
|14.||Bhide SA, Ahmed M, Newbold K, Harrington KJ, Nutting CM. The role of intensity modulated radiotherapy in advanced oral cavity carcinoma. J Cancer Res Ther 2012;8:S67-71. |
|15.||Piotrowski T, Skórska M, Jodda A, Ryczkowski A, Kaz´mierska J, Adamska K, et al. Tomotherapy - Different way of dose delivery in radiotherapy. Wspolcz Onkol 2012;16:16-25. |
|16.||Wu WC, Mui WL. A case report on the effect of fan beam thickness in helical tomotherapy of nasopharyngeal carcinoma. Med Dosim 2011;36:57-61. |
|17.||Kissick MW, Fenwick J, James JA, Jeraj R, Kapatoes JM, Keller H, et al. The helical tomotherapy thread effect. Med Phys 2005;32:1414-23. |
|18.||Kissick MW, Mackie TR, Jeraj R. A delivery transfer function (DTF) analysis for helical tomotherapy. Phys Med Biol 2007;52:2355-65. |
|19.||Shepard DM, Olivera GH, Reckwerdt PJ, Mackie TR. Iterative approaches to dose optimization in tomotherapy. Phys Med Biol 2000;45:69-90. |
|20.||Kinhikar RA, Murthy V, Goel V, Tambe CM, Dhote DS, Deshpande DD. Skin dose measurements using MOSFET and TLD for head and neck patients treated with tomotherapy. Appl Radiat Isot 2009;67:1683-5. |
|21.||Sheng K, Molloy JA, Read PW. Intensity-modulated radiation therapy (IMRT) dosimetry of the head and neck: A comparison of treatment plans using linear accelerator-based IMRT and helical tomotherapy. Int J Radiat Oncol Biol Phys 2006;65:917-23. |
|22.||Fiorino C, Dell′Oca I, Pierelli A, Broggi S, De Martin E, Di Muzio N, et al. Significant improvement in normal tissue sparing and target coverage for head and neck cancer by means of helical tomotherapy. Radiother Oncol 2006;78:276-82. |
|23.||Moldovan M, Fontenot JD, Gibbons JP, Lee TK, Rosen II, Fields RS, et al. Investigation of pitch and jaw width to decrease delivery time of helical tomotherapy treatments for head and neck cancer. Med Dosim 2011;36:397-403. |
|24.||Shrimpton PC, Wall BF, Fisher ES. The tissue-equivalence of the Alderson Rando anthropomorphic phantom for X-rays of diagnostic qualities. Phys Med Biol 1981;26:133-9. |
[Figure 1], [Figure 2], [Figure 3], [Figure 4]
[Table 1], [Table 2], [Table 3], [Table 4]