Chinese-German J Clin Oncol February �0��, Vol. ��, No. �, P89–P9� DOI �0.�007/s�0��0-0��-��70-6

The flexibility in radiotherapy can be improved if a pa-tient can be moved between any one of the department’s medical linear accelerators without the need to change anything in the patient’s treatment plan. For this to be possible, the dosimetric characteristics of the different ac-celerators should be the same i.e. the accelerators must be beam-matched.

Recently two Siemens linear accelerators with 6 and �0 MV photon beams and 6–�� MeV electron beams were installed at our hospital (Children’s Cancer Hospital, Cai-ro, Egypt). The accelerators fulfilled the vendor-defined “fine beam-match” criteria, and a more extensive set of measurements was carried out during commissioning. Performance evaluation of the beam matching from the time of acceptance up tell now were performed. The mea-sured absorbed dose data for the second accelerator were compared with the first accelerator, chosen as reference, and used for the treatment planning system (TPS) calcu-lations.

The beam-matching acceptance criteria defined by the vendor are not strict enough to guarantee optimal beam-match. The safest and most practical way to ensure that the two accelerators are within clinical acceptable accu-racy included TPS calculations in the evaluation. Further, comparisons between measurements and calculations should be done in absolute dose terms [�–�].

The success of curative intended radiation therapy largely depends on the ability to deliver the prescribed dose to the patient within a narrow tolerance. A 5% dif-ference in the delivered dose may result in changes in the order of �0%–�0% in tumor control probability and �0%–�0% in normal tissue complication probability [�]. Thus, there is a need for accurate treatment planning calculations and treatment dose delivery. This imposes narrow tolerance limits on all process of the treatment chain from planning to delivery, including imaging and structure delineation, absorbed dose measurements and calculations, treatment machine performance, and pa-tient setup.

Several publications present recommendations for such tolerance levels [�–5]. They estimates that if the uncertain-ty in dose calculation is between �.0% and 5.0% (� SD) the total uncertainty in the dose delivered to the patient will be between �.0% and 6.�% (� SD) [6]. Wambersie et al recommend a total uncertainty limit of �.5% (� SD) for the dose prescription point although 5% (� SD) is consid-ered more appropriate for other points in the planning target volume. The ICRU �� cited an uncertainty level of ±5%. Accuracy requirement that can be interpreted as an action level were published [7–9]. These recommenda-tions on uncertainty limits and estimations of what can be achieved demonstrate that there is not much room for inaccurate dose calculations. Efforts must be made to re-duce the uncertainty of every link in the treatment chain

Dosimetric evaluation of a beam matching procedureEhab M. Attalla�, �, H. S. Abou-Elenein�, Hany Ammar�, Ismail Eldesoky�

� Children’s Cancer Hospital, Cairo, Egypt � National Cancer Institute, Cairo University, Cairo, Egypt

Received: 10 March 2013 / Revised: 20 October 2013 / Accepted: 20 December 2013© Huazhong University of Science and Technology 2014

Abstract  Objective: The dosimetric characteristics for linear accelerators with the same model, and nominal energy are known to be very similar, as long as the machines are unaltered from the manufacturer’s original specifications. In this pre-liminary study, a quantitative investigation of the similarity in the basic photon and electron dosimetry data from the Siemens Oncor  linear  accelerators  at  our  hospital  (Children’s  Cancer  Hospital,  Cairo,  Egypt)  was  reported.  Methods:  The  output factor (OF), wedge factors (WF), percentage depth dose (PDD), and beam profile for the 6 and 10 MV photon beams were measured. Results: The measured output factors varied by less than about 1% for each field size. The difference between the maximum and minimum PDD values at each depth was less than about 1%. The difference between the beam flattnes and symetry was no more than 1% at all off-axis distances. For electron the results showed that the PDD, OF, and the beam profiles were matched within 1% differences. Conclusion: These results strongly suggest that it is feasible to establish one reference photon and electron dosimetry data set for the two machines and nominal energies.

Key words  beam matching; dosimetric verification; linear accelerator 

Correspondence to: Ehab M. Attalla. Email: h_shaoff@yahoo.com

90 www.springerlink.com/content/�6��-9089

from planning to delivery.Treatment flexibility and efficiency of a radiotherapy

department are very important, and can be greatly im-proved if several linear accelerators can be considered dosimetrically identical and patients can be treated using any linac without the need to adjust the treatment plans. In order to make this possible, the dosimetric characteris-tics of the various accelerators must be the same, or near-ly the same i.e. the accelerators must be beam-matched. Beam-match results for various accelerators have previ-ously been presented [�0]. In addition to beam-matched accelerators only one set of data representing all the ac-celerators needs to be implemented in the TPS.

Materials and methods

The beam-matching procedure was described in detail as per Siemens acceptance test protocol (ATP) [��]. The ac-celerators were purchased with the vendor-defined “fine beam-matching” option. The beam-matching criteria were based on depth ionization curves as well as in-line and cross-line profiles measured in the vendor-defined prescribed geometry. During the customer acceptance procedure (CAP) of the accelerators, all the vendor-de-fined criteria were fulfilled. Even if the measurements were within the vendor-defined criteria adjustments were made to improve the match between the two ac-celerators. Primarily adjustments were made to make both the photon and electron profiles as symmetrical as possible, but for some of the photon and electron beams it was necessary to tune the energy in order to improve

the match of the depth dose curves. Following the CAP more extensive commissioning measurements were un-dertaken (Tables � and �) in view of the facts that: (�) the vendor-defined criteria does not consider full profiles or depth dose curves to evaluate the quality of beam-match-ing but instead only some points on the curves; (�) all vendor-specified measurements for both electrons and photons are done with an ionization chamber and there is no or incorrect ionization chamber offset; (�) the vendor-defined criteria are limited to only a few parameters for a limited set of beam geometries; (�) the beam-matching criteria do not include output factors and therefore a risk of good agreement solely due to normalization may exist; and (5) only open static fields are included in the CAP.

To ensure that the beam-matching was clinically ac-ceptable for photons the commissioning measurements were compared with the TPS, XIO �.0� (Elekta, CMS) superposition beam algorithm calculations. Furthermore, all data were compared in absolute terms, i.e. depth dose curves and profiles for both open and wedged beams were not relatively normalized to the same value at a specific point but to the output factors for each individual acceler-ator. The absolute measurements were performed accord-ing to the TRS �98 protocol [��]. The two linear accelera-tors were prepared and used for IMRT treatments and as a further test of the beam-matching the same IMRT plans were delivered with the two accelerators, measured with the �D array and compared with TPS calculation [��].

For accuracy and precision in radiotherapy the elec-tron beam treatment plans utilized at our clinic were cal-culated using the treatment planning system. Treatment

Table 1  Measurements carried out for photonsMeasurements Equipments SSD (cm) Depth (cm) Field size (cm2)Absolute dosimetry Farmer Chamber PTW 30013, electrometer  100 10 10

  Web line PTW, 3D water phantom PTWTPR 20/10 (Smiflex) 0.125 cc 80 and 90 20 and 10 10Output factors (OF) Farmer Chamber PTW 30013 100 10 4, 5, 6, 8, 10, 12, 15, 20, 25, 30, 35, 40Depth dose Smiflex 0.125 cc 100 From 0 to 35 4, 5, 6, 8, 10, 12, 15, 20, 25, 30, 35, 40Inline and cross line  Smiflex 0.125 cc 100 Dmax, 5, 10, 15, 20 5, 10, 15, 20, 30, 40  profiles open fieldsWedge profiles PWs Smiflex 0.125 cc 100 Dmax, 5, 10, 15, 20 5, 10, 15, 20Wedge profiles VWs Smiflex 0.125 cc 100 Dmax, 5, 10, 15, 20 5, 10, 15, 20Head scatter factors Farmer Chamber PTW 30013 with Perspex buildup 98.5/97.5 for Perspex 1.5/2.5 3, 4, 5, 6, 8, 10, 12, 15,

  cap, and extended SSD for small field sizes   6/10 MV   for 6/10 MV  20, 25, 30, 35, 40

Table 2  Measurements carried out for electronsMeasurements Equipments SSD (cm) Depth (cm) Field size (cm2)Absolute dosimetry Advanced Markus Chamber PTW, electrometer  100 Zref 10

  Web line, 3D Perspex phantom PTWOutput factors (OF) Advanced Markus Chamber PTW 100 Zref 5, 10, 15, 20, 25Depth dose Smiflex 0.125 cc, 3D water phantom PTW 100 – 5, 10, 15, 20, 25Inline and cross line profiles open fields Smiflex 0.125 cc, 3D water phantom PTW 100 Zref 5, 10, 15, 20, 25

9�Chinese-German J Clin Oncol, February �0��, Vol. ��, No. �

planning calculations for electron were compared with that measured at the linear accelerators using �d array.

Results and discussion

Photons measurements The calibration geometry used for absolute dosimetry

was a (�0 × �0) cm� field size, SSD = �00 cm and depth �0 cm for both the 6 and �0 MV beams. The TPR �0/�0 and the percentage depth dose at �0 cm depth (PDD �0) var-ied slightly between the accelerators for the calibration geometry. For both the 6 and �0 MV beams the mean val-ues PDD �0 and TPR �0/�0 of the two accelerators were used for absolute dose calibration. For the 6 MV beam the PDD �0 for the two machines were 66.�% and 66.�% and the TPR�0/�0 for the two machines were 0.666 and 0.670, respectively. For the �0 MV beam the correspond-ing numbers for the PDD �0 were 7�.�% and 7�.�% and the TPR �0/�0 were 0.7�0 and 0.7��, respectively.

Total scatter collimator factorsTotal scatter collimator factors were measured on ON-

COR� and compared with those from ONCOR�. For total scatter collimator factor (TSCF) as shown in Fig. �, the difference between the two accelerators for all field sizes was less than ±0.5%.

Percentage depth dose (PDD)The percentage depth dose (PDD) for the two accelera-

tors were matched within �%, when the measured output factors for the different accelerators were ignored. Fig. � and � showed the measured PDD curves for the 6 and �0 MV photon beams for the two accelerators. From the figures they had showed that there was good matching between the two machines and for quantitative analysis the differences in PDD between the two linear accelera-tors were calculated for all PDD curves at field sizes 5 cm × 5 cm, �0 cm × �0 cm, �5 cm × �5 cm, and �0 cm × �0 cm. The results showed that the maximum difference between the two linear accelerators was less than �% and Fig. � represented the calculated differences between the PDD for �0 MV for the two machines at � different field sizes. As showed in Fig. � the maximum difference was less than �%.

Beam profile measurementsFor the beam profile measurements the comparison

between the two linear accelerators showed that the rela-tive dose profile normalized to �00% at the central axis were matched within a mean value of 0.7% except for field size �0 cm × �0 cm were the deviation between the two accelerators were around �% for both the 6 and �0 MV beams. Fig. 5 and 6 showed the beam profile mea-surements for the two linear accelerators for field size �5 cm × �5 cm at different measurement depths. The com-parison between the two machines, Fig. 7 showed that the maximum difference were �% and this was at the depth of maximum dose (dmax). For the other depths the deviation was less than 0.5%.

The penumbra (�0%–80%) for the two accelerators was determined from the measured profiles for the (�0 × �0) cm� field size. For the 6 MV beams the values were (� ± 0.�) mm for the upper jaws and (5.� ± 0.�) mm for the lower jaws (MLC). For �0 MV the corresponding values were (�.8 ± 0.�) mm and (5.0 ± 0.�) mm.

Fig. 1  The measured total scatter collimator factor (TSCF) in water versus field size for the two LinacsFig. 2  PDD for 6 MV for the two LinacsFig. 3  PDD for 10 MV for the two Linacs

Fig. 4  The differences in PDD between the two linacs for 10 MV at field size 20 cm x 20 cm and 10 cm × 10 cm

9� www.springerlink.com/content/�6��-9089

Wedge factorsWedge factors were calculated by taking ratio of aver-

age of thick end in and thick end out reading to the open field reading. For all field sizes and all wedge angles the differences were less than ±�% for all field sizes. Fig. 8 showed the wedge factors for the two linear accelerators

for field size �0 cm × �0 cm.

Comparison of modulated doses using one data set

A treatment plan was calculated by IMRT using one data set and delivered by the two linear accelerators. The �D dose distribution comparison of a measured and calcu-lated IMRT plan for the two accelerators were presented in Fig. 9. Both the �D dose distribution and profile analy-sis demonstrated that measurements and the calculation agree well and that the accelerators were well matched. The γ 3%, 3 mm analysis using verisoft 3D dose compari-son algorism showed that 99% and 98% of the calculated

Fig. 10  PDD matched curves for  the two Linacs for 21 MeV electron energy for field size 10 cm × 10 cm

Fig. 7  Comparison between measured dose profiles at 15, 50 and 100 mm depths for 6 MV at 15 cm × 15 cm open field

Fig. 8  Wedge factors for the two Linacs at field size 10 cm × 10 cm

Fig. 9  (a) Comparison between measured and calculated 2D dose dis-tributions for 6 MV IMRT plan for Linac 1 measured at 15 cm depth and SSD 85 cm; (b) Comparison between measured and calculated 2D dose distributions  for 6 MV  IMRT plan  for Linac 2 measured at 15 cm depth and SSD 85 cm

Fig. 6  Beam profile in cross plan for the two linear accelerators for 6 MV for field size 15 cm × 15 cm at different depths

Fig. 5  Beam profile for in-plan axis for the two linear accelerators for 10 MV for field size 15 cm × 15 cm at different depths

Fig. 11  Matched beam profiles for 21 MeV electron energy for applica-tor 25 cm × 25 cm at Zref

9�Chinese-German J Clin Oncol, February �0��, Vol. ��, No. �

points of the IMRT plan were succeeded the γ index cri-teria and the failed points at the same points in the low dose region for the two accelerators.

Electrons For percentage depth dose measurements in electron

the results showed that the maximum deviation (MRp, MR50, MR80, MR85, and MR90) was 0.7 mm for all electron energies and applicators when comparing the depths of the 90% (R90), 85% (R85) , 80% (R80) and the depth of the 50% dose (R50) and the practical range (Rp) with the mean values (Table �). Fig. �0 showed PDD matched curves for the two Linacs for �� MeV electron energy for field size �0 cm × �0 cm. Also the output factors at dmax were in good agreement where the maximum deviation was less than 0.5%. In addition, the profiles agreed within a clinically acceptable level. Better agreement was gen-erally found for smaller applicators (Fig. ��). The largest deviation between profiles for the two accelerators was less than �%, and the agreement was generally within �% for applicators smaller than �0 cm × �0 cm.

ConclusionThis work was done to evaluate the extent of beam

matching between two Siemens ONCOR expression ac-celerators. The two machines data sets can be represented by one data set and implemented to the TPS. The calcu-lated treatment plans can be delivered by any accelerator and this will facilitate for patients and planners.

References

Bhangle JR, Narayanan VK, Kumar NK, et al. Dosimetric analysis of beam-matching procedure of  two similar  linear accelerators. J Med Phys, 2011, 36: 176–180. Sjöström D, Bjelkengren U, Ottosson W, et al. A beam-matching con-cept for medical linear accelerators. Acta Oncol, 2009, 48: 192–200. Chetty IJ, Curran B, Cygler JE, et al. Report of the AAPM Task Group No.  105:  Issues  associated  with  clinical  implementation  of  Monte Carlo-based photon and electron external beam treatment planning. Med Phys, 2007, 34: 4818–4853.Mijnheer B, Olszewska A, Fiorino C, et al. Quality assurance of treat-ment  planning  systems.  Practical  examples  for  non-IMRT  photon beams. Estro: Booklet 7, 2004.Venselaar J, Welleweerd H, Mijnheer B. Tolerances for the accuracy of photon beam dose calculations of treatment planning systems. Ra-diother Oncol, 2001, 60: 191–201. Ahnesjö A, Aspradakis  MM.  Dose  calculations  for  external  photon beams in radiotherapy. Phys Med Biol, 1999, 44: R99–R155. Goitein M. Nonstandard deviations. Med Phys, 1983, 10: 709–711. Goitein M. Calculation of the uncertainty in the dose delivered during radiation therapy. Med Phys, 1985, 12: 608–612. Hrbacek J, Depuydt T, Nulens A, et al. Quantitative evaluation of a beam-matching procedure using one-dimensional gamma analysis. Med Phys, 2007, 34: 2917–2927.Watts RJ. Comparative measurements on a series of accelerators by the same vendor. Med Phys, 1999, 26: 2581–2585. Klein EE, Hanley J, Bayouth J, et al. Task Group142 report: quality as-surance of medical accelerators. Med Phys, 2009, 36: 4197–4212.  IAEA TRS-398. Absorbed  dose  determination  in  external  beam  ra-diotherapy: an international code of practice for dosimetry based on standards of absorbed dose to water. IAEA, 2000. Low DA, Harms WB, Mutic S, et al. A technique for the quantitative evaluation of dose distributions. Med Phys, 1998, 25: 656–661.

1.

2.

3.

4.

5.

6.

7.8.

9.

10.

11.

12.

13.

Table 3  The maximum deviation (MRp,MR50, MR80, MR85, MR90, and MOF) between the two accelerators

Value Linac 16 MeV

Linac 2 6 MeV

Linac 1 9 MeV

Linac 2 9 MeV

Linac 112 MeV

Linac 212 MeV

Linac 115 MeV

Linac 215 MeV

Linac 118 MeV

Linac 218 MeV

Linac 121 MeV

Linac 221 MeV

Dmax (mm) 12 11.9 19 19.1 26 25.99 24 24 14 14 11 11R50 (mm) 22.02 21.99 33.98 33.99 46.01 46.02 57.94 57.94 73.27 73.27 83.77 83.07R90 (mm) 16.27 16.27 25.77 25.77 35.09 35.09 43.05 43.11 50.21 50.21 51.99 51.99R85 (mm) 17.29 17.27 27.22 27.23 36.93 36.95 46.05 46.03 55.43 55.43 59.34 59.34R80 (mm) 18.1 18.09 28.48 28.45 38.69 38.63 48.29 48.29 59.24 59.24 64.74 64.74Rp (mm) 28.19 28.18 42.3 42.3 56.6 56.7 71.03 71.09 90.82 90.83 104.06 104.11