Quantification of the role of lead foil in flattening filter free beam reference dosimetry

Abstract Purpose To quantify the potential error in outputs for flattening filter free (FFF) beams associated with use of a lead foil in beam quality determination per the addendum protocol for TG‐51, we examined differences in measurements of the beam quality conversion factor kQ when using or not using lead foil. Methods Two FFF beams, a 6 MV FFF and a 10 MV FFF, were calibrated on eight Varian TrueBeams and two Elekta Versa HD linear accelerators (linacs) according to the TG‐51 addendum protocol by using Farmer ionization chambers [TN 30013 (PTW) and SNC600c (Sun Nuclear)] with traceable absorbed dose‐to‐water calibrations. In determining kQ, the percentage depth‐dose at 10 cm [PDD(10)] was measured with 10×10 cm2 field size at 100 cm source‐to‐surface distance (SSD). PDD(10) values were measured either with a 1 mm lead foil positioned in the path of the beam [%dd(10)Pb] or with omission of a lead foil [%dd(10)]. The %dd(10)x values were then calculated and the kQ factors determined by using the empirical fit equation in the TG‐51 addendum for the PTW 30013 chambers. A similar equation was used to calculate kQ for the SNC600c chamber, with the fitting parameters taken from a very recent Monte Carlo study. The differences in kQ factors were compared for with lead foil vs. without lead foil. Results Differences in %dd(10)x with lead foil and with omission of lead foil were 0.9 ± 0.2% for the 6 MV FFF beam and 0.6 ± 0.1% for the 10 MV FFF beam. Differences in kQ values with lead foil and with omission of lead foil were −0.1 ± 0.02% for the 6 MV FFF and −0.1 ± 0.01% for the 10 MV FFF beams. Conclusion With evaluation of the lead foil role in determination of the kQ factor for FFF beams. Our results suggest that the omission of lead foil introduces approximately 0.1% of error for reference dosimetry of FFF beams on both TrueBeam and Versa platforms.

the flattening filter is removed from the beam path, the FFF beams have higher dose per pulse (DPP) and more forward peaked bremsstrahlung than the flattened beam.The process for measuring absolute dose for flattened beams has been outlined in the AAPM TG-51 1 protocol.An addendum 2 to TG-51 includes guidance for FFF beams to address ionization recombination effects due to high DPP and volume averaging effects from unflattened beam profiles.Comprehensive evaluations of these two effects were previously published. 3,4To determine the beam quality conversion factor k Q , TG-51 1 recommends using a 1 mm lead foil for beam energies higher than 10 MV to remove unknown electron contamination from the linac head.TG-51 addendum 2 no longer recommends the use of lead foil for flattened high-energy photon beams.A comprehensive study 5 showed that omitting lead foil led to errors in k Q of no more than 0.2% and simplified the measurement procedure, making it less prone to operator error.The addendum 2 still recommends use of a 1 mm lead foil for FFF beams owing to a lack of data.A comprehensive comparison between recommendations of the TG-51 and TG-51 addendum reference dosimetry protocols 4 demonstrated that differences in k Q determined using the two protocols for PTW N30013 ionization chambers were within 0.1% for FFF beams.For the TG-51 addendum protocol, they also considered corrections for defined radiation field size, ion chamber leakage, and volume averaging effects in the output measurements.Another investigation of the effect of including lead foil in the beam quality measurement for 6 MV FFF and 10 MV FFF beams 6 showed that the change in k Q was 0.1% on a Varian TrueBeam linac for the PTW 30013 Farmer chamber.That group included ion recombination and beam non-uniformity corrections in percent depth dose at 10 cm, PDD (10), measurements obtained with lead foil, but those two corrections have not been considered for PDD (10) measurements with the omission of lead foil.A very recent publication of guidance for TG-51 reference dosimetry 7 recommends that the use of lead foil is required for high accuracy determination of beam quality specifiers for photon beams with energy of 10 MV and higher based on the previously studies. 5he purpose of this study was to quantify the differences in the beam quality conversion factor k Q between the use of lead foil and with omission of use lead foil for FFF beams under the same measurement conditions.From these differences in k Q , measured on several True-Beam and Versa HD linacs, the error in output resulting from omitting the lead foil was determined.

METHODS AND MATERIALS
We measured the output a 6 MV FFF and a 10 MV FFF beam on eight Varian TrueBeam linacs (Varian Medical Systems, Palo Alto, California, USA), and two Elekta Versa HD linacs (Elekta Solutions AB, Stockholm, Sweden).The measurements were conducted in a 1D water tank by using Farmer ionization chambers with traceable absorbed dose-to-water calibrations and The setup for beam quality measurements involving lead foil.
PC electrometer (Sun Nuclear, Melbourne FL), also with traceable calibrations.Two types of waterproof Farmer ionization chambers were used, a PTW 30013 with an active volume of 0.6 cm 3 (PTW, Freiburg, Germany) and an SNC600c also with an active volume of 0.6 cm 3 (Sun Nuclear, Melbourne FL).The PDD(10) measurements were made both with a lead foil and without a lead foil in place.All PDD(10) measurements were done with a 10 × 10 cm 2 field size at 100 cm source-to-surface distance (SSD), and the data was shifted to the effective point of measurement of the ion chamber as specified in the TG-51 report. 1 The measurements with the lead foil [%dd (10) Pb ] and without the lead foil [%dd (10)] were acquired under the same setup conditions.When lead foil was used, it was placed in the exit window of the machine head, approximately 50 cm from the water surface (Figure 1).When the lead foil was used, the photon component of the PDD (10), %dd (10) x , was calculated from the %dd(10) Pb according to eq. ( 13) in TG-51 1 : if %dd(10) Pb ≥ 73%; otherwise, %dd (10) x = %dd(10) Pb .
When lead foil was omitted, %dd (10) x , was determined directly from %dd (10), %dd (10) x = %dd (10), for both 6 MV FFF and 10 MV FFF beams.It should be noted that for the 10 MV FFF beam on the Versa in this study the %dd (10) Pb required the use of the The measured %dd (10) and the resultant %dd(10) X , and k Q with omission of use lead foil were compared with those measurements made with a 1-mm lead foil for the two individual linacs as an example.
above empirical formula but the %dd (10) was below the threshold required for the use of the interim alternative formula in TG-51. 1 The differences in %dd (10) x with and without lead foil were compared for the measurement data from the ten linacs.The beam quality correction factor k Q then was determined by using the empirical fit equation ( 1) in the TG-51 addendum 2 : 63 < %dd (10) x < 86.
The fitting parameters, a = 0.9562, b = 2.141, and c = −2.623,for the PTW 30013 ionization chamber were from Table 1 in the TG-51 addendum, 2 whereas the fitting parameters for the SNC600c ionization chamber were from Table 6 of the recent Monte Carlo study. 8he SNC600c ionization chamber was used for Linac 1 and Linac 2, and the PTW 30013 ionization chamber were used for the other linacs.Two metrics were chosen to compare the effects of measurements with lead foil and with the omission of lead foil, %dd (10) x and k Q. .The change in %dd (10) x (with foil) was quantified by Δ[%dd (10) x ] = %dd (10) x [lead]-%dd (10) x [no lead]; and the change in k Q by −1 for measurement data from 10 linacs.The output error due to the differences in k Q factors was thus analyzed.
The effect of including the lead foil in the beam quality measurement %dd (10) x , and the k Q factors for all 10 linacs in this study (Figure 2a), indicate that the average differences in %dd (10) x measured with lead foil and with omission of lead foil were 0.9 ± 0.2% for the 6 MV FFF beam and 0.6 ± 0.1% for the 10 MV FFF beam.The maximum differences in %dd (10) x with lead foil and with omission of lead foil were 1.2% for the 6 MV FFF beam and 0.8% for the 10 MV FFF beam.When lead foil was used, the %dd (10)x value was always higher than that without lead foil for the same energy beam, because the electron contamination is reduced by using the lead foil, which is more pronounced at d max than that at 10 cm depth.
Measurement data for all 10 linacs in this study (Figure 2b) also showed that the average differences in k Q factor measured with and without lead foil were −0.1 ± 0.02% with a maximum of −0.13% for the 6 MV FFF beam, and −0.1 ± 0.01% with a maximum of −0.12% for the 10 MV FFF beam.Those results demonstrated that when calibrating FFF beams, the added uncertainty from the omission of using lead foil in the beam quality determinations is about 0.1%.Therefore, the error in output determination resulting from omitting of using lead foil is also about 0.1%.

DISCUSSION
To correct for electron contamination in photon beams at d max during reference dosimetry, Li and Rogers concluded from Monte Carlo studies that placing a 1-mm lead foil directly below the linac head yielded the best results. 9,10The AAPM TG-51 protocol recommends using 1 mm lead foil for determining the photon beamquality specifier %dd(10) X for high-energy beams above 10 MV, with the lead foil placed at 50 or 30 cm from the water surface.The AAPM TG-51 addendum recommended using 1 mm lead foil only for FFF beams for all beam energies. 2The k Q results we obtained for 6 MV FFF and 10 MV FFF beams from Varian True-Beam and Elekta Versa HD linacs showed no significant differences for determining k Q when Farmer-type ionization chambers were used.Similar studies on Halcyon linacs for 6 MV FFF beams also indicated that machine output values differed by only 0.1% between the beam quality measurements without versus with 1 mm lead foil. 11,12Our results were consistent with previous studies showing that the change in k Q from including lead foil in the beam was sufficiently small compared with the beam quality data obtained without lead foil in the beam for reference dosimetry of both 6 MV FFF and 10 MV FFF beams on TrueBeam and Versa platforms with Farmer-type ionization chambers.We did not include the effect of the variation in P ion between d max and 10 cm depth in k Q determination.The variation of P ion with depth is chamber dependent and is larger for FFF beams than for flattened beams.We found that for the PTW 30013 (Farmer) chamber the variation of P ion between d max and 10 cm was 0.3% in 6 MV FFF beam and 0.2% in 10 MV FFF beam, which is consistent with previous studies. 13The P ion variation of this chamber adds a maximum 0.03% uncertainty in k Q due to the shallow slope of the k Q versus PDD (10) X curve.This result may not apply for other type chambers and/or bias voltages and should be studied by each user with their equipment.

CONCLUSION
The recommendations of the TG-51 guidances 2,7 require the use of lead foil in k Q determination of FFF beam reference dosimetry.Our results of 6 MV FFF and 10 MV FFF beams from both TrueBeam and Versa linacs suggest that a 0.1% of error introduced in FFF beam reference dosimetry from the omission of lead foil.

F I G U R E 2
Effect of using lead foil on 6 MV FFF and 10 MV FFF beam quality measurements on 10 linacs.(a) Differences in %dd(10)x for no lead foil vs with lead foil.(b) Differences in the k Q factor for no lead foil versus with lead foil.Linacs #1 and #2 are TrueBeam linacs measured with an SNC600c ionization chamber.Linacs #9 and #10 are Versa HD linacs measured with a PTW ionization chamber.The others linacs are TrueBeam machines measured with PTW ionization chambers.
Effect of using lead foil on 6 MV FFF and 10 MV FFF beam quality measurements TA B L E 1