If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding authors at: Institut Curie, PSL Research University, Radiation Oncology Department, Proton Therapy Center, Centre Universitaire, 91898 Orsay, France.
Institut Curie, PSL Research University, Université Paris-Saclay, CNRS UMR 3347, INSERM U1021, 91898 Orsay, FranceInstitut Curie, PSL Research University, Radiation Oncology Department, Proton Therapy Center, Centre Universitaire, 91898 Orsay, France
Institut de Radioprotection et de Sûreté Nucléaire, Service de Dosimétrie, Laboratoire de Dosimétrie des Rayonnements Ionisants, 92262 Fontenay-aux-Roses Cedex, France
Institut de Radioprotection et de Sûreté Nucléaire, Service de Dosimétrie, Laboratoire de Dosimétrie des Rayonnements Ionisants, 92262 Fontenay-aux-Roses Cedex, France
Corresponding authors at: Institut Curie, PSL Research University, Radiation Oncology Department, Proton Therapy Center, Centre Universitaire, 91898 Orsay, France.
Characterization of set-up capable of proton FLASH irradiations.
•
Mean dose rates of up to 385 Gy/s and instantaneous dose rates of 1000 Gy/s were obtained.
•
Evaluation of ion collection efficiency of several ionization chambers in very-high dose rates.
•
Comparison of methods employed in the calculation of ion recombination coefficients.
•
Reduction in collection efficiency for very-high dose rates cannot be ignored in PBS FLASH experiments.
Abstract
Purpose
Accurate dosimetry is paramount to study the FLASH biological effect since dose and dose rate are critical dosimetric parameters governing its underlying mechanisms. With the goal of assessing the suitability of standard clinical dosimeters in a very-high dose rate (VHDR) experimental setup, we evaluated the ion collection efficiency of several commercially available air-vented ionization chambers (IC) in conventional and VHDR proton irradiation conditions.
Methods
A cyclotron at the Orsay Proton Therapy Center was used to deliver VHDR pencil beam scanning irradiation. Ion recombination correction factors (ks) were determined for several detectors (Advanced Markus, PPC05, Nano Razor, CC01) at the entrance of the plateau and at the Bragg peak, using the Niatel model, the Two-voltage method and Boag’s analytical formula for continuous beams.
Results
Mean dose rates ranged from 4 Gy/s to 385 Gy/s, and instantaneous dose rates up to 1000 Gy/s were obtained with the experimental set-up. Recombination correction factors below 2 % were obtained for all chambers, except for the Nano Razor, at VHDRs with variations among detectors, while ks values were significantly smaller (0.8 %) for conventional dose rates.
Conclusions
While the collection efficiency of the probed ICs in scanned VHDR proton therapy is comparable to those in the conventional regime with recombination coefficiens smaller than 1 % for mean dose rates up to 177 Gy/s, the reduction in collection efficiency for higher dose rates cannot be ignored when measuring the absorbed dose in pre-clinical proton scanned FLASH experiments and clinical trials.
Radiotherapy is a curative treatment for a large variety of cancers and is one of the key components of effective cancer treatment, comprising 50 % to 60 % of cancer treatments. It operates on a thin line between maximizing the eradication of tumour cells while minimizing the damage to nearby healthy tissues. Already observed in 1959 by Dewey and Boag [
] who used an electron accelerator to produce mean dose rate electron beams greater than 40 Gy/s, and observed that FLASH-irradiated mice did not exhibit pneumonitis and lung fibrosis, unlike conventional dose rate (≤0.03 Gy/s) irradiated mice, while retaining tumour control – FLASH effect. FLASH radiation therapy (FLASH-RT) may open the door to an increase of the therapeutic index, allowing for dose escalation especially relevant to radio-resistant tumours, such as glioblastoma, and to potentially circumvent the motion-induced uncertainties of the delivered radiation, given that the time scale of the irradiation itself is much shorter than any patient’s movement. In FLASH-RT the therapeutic dose is delivered within a few milliseconds (Very-High dose rate or VHDR) or microseconds (Ultra-High dose rate or UHDR), while in conventional radiation therapy (CONV-RT) irradiations take a few minutes (in what follows, we will use this convention for the definition of dose rates). Since the first seminal work of Favaudon et al. the FLASH effect has been confirmed across multiple tissue types [
Fouillade C, Curras-Alonso S, Giuranno L, Quelennec E, Heinrich S, Bonnet-Boissinot S, et al. FLASH irradiation spares lung progenitor cells and limits the incidence of radio-induced senescence. Clin Cancer Res 2020;26:1497–506. https://doi.org/10.1158/1078-0432.CCR-19-1440.
Reduced cognitive deficits after FLASH irradiation of whole mouse brain are associated with less hippocampal dendritic spine loss and neuroinflammation.
Montay-Gruel P, Acharya MM, Jorge PG, Petit B, Petridis IG, Fuchs P, et al. Hypofractionated FLASH-RT as an Effective Treatment against Glioblastoma that Reduces Neurocognitive Side Effects in Mice. Clin Cancer Res 2021;27:775–84. https://doi.org/10.1158/1078-0432.CCR-20-0894.
Feasibility Study of FLASH Radiotherapy for the Treatment of Symptomatic Bone Metastases ClinicalTrials. https://clinicaltrials.gov/ct2/show/NCT04592887.
] is underway in the USA to investigate the feasibility of proton FLASH-RT for the palliative treatment of painful bone metastases.
Clinical proton therapy beamlines can be modified to be capable of (pre-) clinical studies in the FLASH regime, with many proton therapy centers already reporting the successful delivery of FLASH proton beams either with cyclotrons [
]. For that reason, it is paramount to characterize the already available dosimetric tools in the conditions of proton FLASH dose rates and envision possible technological upgrades [
], although at the present moment no recommendations or dosimetric protocols exist for FLASH-RT. Ionization chambers exhibit a reduction of the charge collection efficiency due to the recombination of ions along the original ionization track before reaching the collecting electrodes – initial recombination [
]. This effect is more predominant in high linear energy transfer beams or regions at the distal end of the Bragg peak, in which the ionization density is higher. In addition, ions belonging to different ionization tracks can combine with each other, thereby reducing the charge collected by the ionization chamber - volume recombination [
]. This mechanism is highly dependent on the ionization current and, consequently, on the dose rate, and in FLASH dose rates it becomes the dominant effect that leads to the reduction of the collected charge. The recombination coefficient, ks - inverse of the charge collection efficiency, needs to be determined for each chamber as it is voltage and geometry-dependent.
Substantial work has already been carried out on pulsed electron FLASH dosimetry. However, current theoretical models [
], does not take into account the fraction of free electrons that are collected by the electrodes without inducing any ionization, which can result in the underestimation of the ion collection efficiency in high DPP beams. In narrow chambers, operated at high voltages and in high DPP beams, the space charge results in a distortion of the electric field leading to a reduction of the field strength inside the ionization chamber, and consequently to a greater ionic recombination. Petersson et al. [
] proposed an empirical model to correct the ion recombination in the Advanced Markus chamber in high DPP electron beams (10−4 Gy–10 Gy) for recommended polarization voltages. The ion collection efficiency was shown to decrease with increasing DPP, while remaining unaltered with the dose rate, and ks values were higher than 3. Gotz et al. [
] developed a numerical model of the transport of charges in a plane-parallel ionization chamber that includes the effect of the charges and their inhomogeneous distribution on the chamber’s electric field. Furthermore, the model showed a better agreement with measurements done with a PTW Advanced Markus chamber at higher collection bias, than Boag’s theory and was in agreement with the empirical model of Petersson et al. [
The dose rate in conventional proton pencil beam scanning (PBS), with instantaneous dose rates of a few Gy/s, is much higher than in scattered beams, which can therefore significantly modify the charge collection and recombination factors expected in ionization chambers. The ion recombination effect has already been evaluated in scanned beams by several groups employing different methods. Liszka et al. [
] evaluated the ion recombination of an air-vented ionization chamber in conventional continuous PBS beams and observed a dependency on the beam energy and dose rate, the two being correlated, concluding that volume recombination should not be disregarded and that recombination coefficients ought to be determined for each employed chamber and beam quality. Rossome et al. [
Rossomme S, Delor A, Lorentini S, Vidal M, Brons S, Jkel O, et al. Three-voltage linear method to determine ion recombination in proton and light-ion beams. Phys Med Biol 2020;65:045015. https://doi.org/10.1088/1361-6560/ab3779.
] have proposed an alternative method to the TVM for protons and light-ion beams, using three current measurements at three different polarizing voltages (V), to determine the saturation current, provided that the linearity of ks with (pulsed beams) and 1/ (continuous beams) can be established. This method, appropriately called the three-voltage linear method, was shown to be more accurate than the TVM since the use of three voltages ensures the verification of the previously mentioned linearity. Yin et al. [
] compared the TVM and the three-voltage method to theoretical values of ion recombination coefficients in passive scattering proton FLASH conditions. A good agreement was observed among the three methods for parallel-plate chambers for mean dose rates of up to 127.58 Gy/s; however, they concluded that cylindrical chambers, such as the IBA CC13 chamber or larger chambers, should not be employed in proton FLASH dosimetry as ion recombination correction values were larger than 50 % for dose rates ≥ 63.71 Gy/s. Cunningham et al. [
], we have developed an experimental set-up capable of performing FLASH-PT irradiations, based on scattered beams. We have characterized this system using dosimetric methods and compared several detectors, for mean dose rates up to 80 Gy/s. Motivated by the growing evidence and support for FLASH, we report in this work the characterisation of a new set-up dedicated to pre-clinical experiments and based on scanned beams, capable of VHDR irradiations with instantaneous dose rates of 1000 Gy/s. Scarce data or experimental set-ups exist at these dose rate conditions, or at least for this configuration of a clinical machine, so we propose an evaluation of the recombination factors for a set of detectors with different geometries, some of which commonly used in proton therapy, and the comparison of experimental results with theoretical models.
2. Materials and methods
In this work we studied the charge collection efficiency of several air-vented ionization chambers with different geometries (at different depths in water) in VHDR beams, and compared them to the conventional irradiation regime. Herein, we define mean dose rate as the total dose over the total deliver time; instantaneous dose rate as the dose rate of a single pencil beam; and since each pencil beam contributes with a different dose to the total dose measured in an ionization chamber, we define the mean effective dose rate, , using the following expression[
where is the dose of the -th pencil beam with instantaneous dose rate .
The evaluated chambers were the PPC05, the CC01 and the Razor Nano-chamber CC003 (IBA Dosimetry, Schwarzenbruck, Germany), and two PTW Advanced Markus type 34,045 (PTW, Freiburg im Breisgau, Germany) with series number (SN) 870 and 879. Their characteristics are summarised in Table 1. We also estimated the response of the PTW microDiamond Type 60019, detector, which is indicated for small field dosimetry, and the IBA Razor Diode detector, with no external bias voltage applied, as their response to such dose rate levels has never been reported.
Table 1Properties of the ionization chambers and semiconductor detectors investigated in this work. Ionization chamber quantities were obtained from vendor websites (https://www.ptwdosimetry.com, https://www.iba-dosimetry.com) unless a literature reference is given.
Rossomme S, Delor A, Lorentini S, Vidal M, Brons S, Jkel O, et al. Three-voltage linear method to determine ion recombination in proton and light-ion beams. Phys Med Biol 2020;65:045015. https://doi.org/10.1088/1361-6560/ab3779.
where the term A/ corresponds to initial recombination and B/ accounts for volume recombination effects. The charge saturation value,, is then obtained by fitting the curves of the inverse of the collected charge, 1/, versus the inverse of the square of the chambers’ polarizing voltage, 1/, for continuous beams with the polynomial
(3)
where and = . In addition, the TVM can also be applied to estimate the volume recombination when initial recombination can be ignored, which for continuous beams follows the expression:
(4)
where is the polarizing voltage used to measure the charge . In this work, we used = 300 V and = 100 V. This method is based on Boag’s model and is applicable in the linear regime of the inverse of the charge with the inverse of the applied voltage to the chamber for pulsed beams, and the inverse of the square of the applied voltage for continuous beams [
]. Both CONV and VHDR proton pencil beams can be considered continuous, as the bunch duration was much shorter than the considered chambers’ collection time (few μs) and the RF frequency of the cyclotron was 106 MHz [
] (3.97 1014 sm-1C-1V2), is a parameter dependent on the geometry of the chamber, is the ionization chamber’s volume and the applied bias. The mean effective current,, was calculated considering the individual contribution of each beam spot to the total current measured by the chamber, as well as the spot’s duration, similarly to eq. (1).
2.1 Experimental set-up
The Proton Therapy Center at Institut Curie houses one “universal” nozzle-equipped gantry supplied by a Proteus 235 isochronous cyclotron (IBA, Belgium) capable of delivering both pencil beam scanning and double scattering (DS) treatment modalities and also very-high dose rate irradiation. This model has already been used by several groups to perform FLASH irradiations using different experimental set-ups, for example with single scattered [
] or scanned beams. In our VHDR set-up with scanned beams, the dose rate was maximized by using the maximum nominal cyclotron current of 500 nA and 226.899 MeV energy, as any energy degrader introduces substantial losses, besides the optimization of the beam transport. The pencil beam scanning, performed with two scanning magnets, was also optimised to reduce the dead time between each pencil beam position. With this specific set-up, also used for radiobiology experiments, an initially accelerated narrow monoenergetic proton pencil beam (in our case the spatial spread of the pencil beam is 4.0 ± 0.3 mm in air at isocenter) can be scanned over a 15 × 15 mm2 square field (a uniform broad field is generated by the superposition of a finite number of equally-spaced proton pencil beams), thus satisfying a priori requirements of VHDR irradiation for healthy tissues (the constraint that we have adopted is that the irradiation duration has to be shorter than 100 ms and mean dose rates are greater than 100 Gy/s). The electrometer of the monitoring ionization chambers was adapted with a resistive divider to ensure its operation in VHDR mode, and the delivery system control files [
Toscano S, Souris K, Gomà C, Barragán-Montero A, Puydupin S, Stappen F vander, et al. Impact of machine log-files uncertainties on the quality assurance of proton pencil beam scanning treatment delivery. Phys Med Biol 2019;64:095021. https://doi.org/10.1088/1361-6560/AB120C.
] (or log files) were used to characterize the time structure of the VHDR beams, necessary to estimate instantaneous and mean dose rates. The read-out of the monitoring ionization chambers was performed every 250 μs, which is shorter than the delivery time of each pencil beam (the minimum spot duration was set to 1 ms). In this configuration each pencil beam had a duration of approximately 7 ms with a total irradiation time of 99 ms for FLASH irradiations of 17.5 Gy. A description of the experimental set-up is shown in Fig. 1, in which one can see the distribution of the spots, as well as the way the dose varies spatially and over time during the irradiation. The physical properties of the irradiation fields used in this work are summarized in Table 2. In particular, we can see in this figure that the spatial dose rate distribution in a PBS plan depends on the arrangement and the distance between the pencil beams. When the dose rate dependence of the detector response is no longer negligible, and especially if the detector is small, it may then be necessary to take into account the effective dose rate at the measurement point in order to correctly evaluate the detector response.
Fig. 1Description of the experimental setup (plateau region of the Bragg peak curve, 2 cm depth in water): (a) 2D dose distribution of all spots, (b) corresponding dose rate distribution according to equation (1). (c) Cumulated and instantaneous dose rate as a function of time at the center of the field and (d) central dose profiles measured with the 2D scintillator detector Lynx (IBA Dosimetry).
Throughout this work, reference dosimetry was performed with the Advanced Markus chamber (SN 870) calibrated under reference conditions at the PTW-Freiburg calibration laboratory (Freiburg, Germany) in terms of absorbed dose-to-water. We applied the calibration coefficient of 1.4466 Gy/nC, the polarity correction factor of 1 [
Gomà C, Sterpin E. Monte Carlo calculation of beam quality correction factors in proton beams using PENH. Phys Med Biol 2019;64:185009. https://doi.org/10.1088/1361-6560/AB3B94.
], the ion recombination correction obtained as in section 2.1 and the temperature and pressure correction factor.
Ion collection efficiencies for CONV and VHDR irradiations were evaluated at the plateau region of the Bragg peak curve (2 cm depth in water), corresponding to a mean dose rate measured with the Advanced Markus chamber (SN 870) of 4.1 Gy/s (7.0 Gy/s mean effective dose rate) and 177 Gy/s (437 Gy/s mean effective dose rate), respectively, and at the Bragg Peak (29.4 cm depth in water), corresponding to 385 Gy/s in the VHDR mode (951 Gy/s mean effective dose rate). The IBA BluePhantom was used for precise positioning of the ionization chambers. Recombination coefficients were also evaluated for different instantaneous dose rates, corresponding to the contributions of the different pencil beams to the total dose of the irradiated field measured with the Advanced Markus chamber (SN 870): this was achieved by measuring the dose delivered by each of the pencil beams composing the square field, irradiating them one by one, sequentially. This procedure was repeated for four different polarizing voltages and Niatel’s model was applied to obtain the coefficients. The IBA Dose1 electrometer (IBA-Dosimetry, Schwarzenbruck, Germany) was used to read the charge collected by the ionization chambers and to apply the polarizing voltage. We measured the integrated charge for different polarizing voltages: 100 V, 150 V, 200 V, 300 V, 400 V and recorded five measurements for each voltage.
3. Results
Three methods were applied to calculate the ion recombination coefficients of five ionization chambers with different dimensions and geometries: Niatel’s model, the TVM and Boag’s analytical formula. Fig. 2a shows normalized plots of 1/ as a function of 1/ obtained in CONV irradiation conditions measured at the plateau of the Bragg peak curve, while Fig. 3, Fig. 4 show the equivalent plots in VHDR conditions, both measured at the plateau and at the Bragg peak. The error bars correspond to 1 standard deviation of the mean. The plots were fitted with a linear fit (Fig. 3) and with a polynomial fit corresponding to eq. (3) (Fig. 4). In both CONV and VHDR conditions the data points at the polarizing voltage of 400 V were not considered in the fit of both Advanced Markus chambers, since it appears to be a distortion due to the effect of charge multiplication, confirmed by plotting 1/ versus 1/.
Fig. 2(a) The inverse of the collected charge, 1/, normalized to the collected charge with the ionization chamber biased to 300 V, , as a function of the inverse of the chamber’s polarizing voltage, 1/, for the Advanced Markus SN 870 and SN 879, the PPC05, the CC01 and the Nano Razor CC003 chamber. The measurements were performed in CONV conditions at the plateau of the Bragg peak. The lines correspond to polynomial fits as in eq. (3), respective R2 are shown. (b) ks values as a function of the inverse of the ionization chambers’ bias, 1/, obtained with the polynomial fit, eq. (3). The curves correspond to the Niatel model, eq. (2) using the fitting parameteres of Table 3. The error bars correspond to statistical uncertainties of 1 σ. The data point for the polarization voltage 400 V was not considered in the fit for the Advanced Markus chamber SN 870 and SN 879.
Fig. 3The inverse of the collected charge, 1/, normalized to the collected charge with the ionization chamber biased to 300 V, , as a function of the inverse of the chamber’s polarizing voltage, 1/, for the Advanced Markus SN 870 and SN 879, the PPC05, the CC01 and the Nano Razor CC003 chamber. The measurements in (a) were performed at the plateau of the Bragg peak, while in (b) the measurements at the Bragg peak are shown, both in VHDR conditions. The lines correspond to linear fits, respective R2 are shown. The error bars correspond to statistical uncertainties of 1 σ. The data point for the polarization voltage 400 V was not considered in the fit for the Advanced Markus chamber SN 870 and SN 879.
The fitting parameters of the polynomial fits (obtained from non-normalized plots) are presented in Table 3, along with 95 % confidence intervals evaluated with the Student’s t-distribution. Table 4 summarizes the ion recombination coefficients determined for CONV (4.1 Gy/s) and VHDR irradiations, the latter includes coefficients evaluated at the entrance plateau, corresponding to the mean dose rate of 177 Gy/s, and at the Bragg peak, corresponding to the mean dose rate of 385 Gy/s. Fig. 2b shows ks, calculated with eq. (2), as a function of the inverse of the polarizing voltage for CONV conditions, the equivalent is presented in Fig. 5 for VHDR conditions.
Table 3Fitting parameters of Niatel’s model, eq. 3, along with 95 % confidence intervals evaluated with the Student’s t-distribution, for five ionization chambers in CONV (4.1 Gy/s) and VHDR (177 Gy/s and 385 Gy/s) irradiation modalities.
Table 4Recombination coefficients, for the bias voltage of 300 V, calculated with Niatel’s model, the TVM and with Boag’s analytical formula, for five ionization chambers in CONV (4.1 Gy/s) and VHDR (177 Gy/s and 385 Gy/s) irradiation modalities. Percentage differences between the Niatel model and TVM or Boag are shown in parenthesis.
Fig. 5ks values as a function of the inverse of the ionization chambers’s bias, 1/, obtained with the polynomial fit, eq. (3), at the plateau (a) and at the Bragg peak (b). The curves correspond to the Niatel model, eq. (2), using the fitting parameters. The error bars correspond to statistical uncertainties of 1σ.
In Fig. 6a, one can observe that the IBA PPC05 chamber exhibits the highest collection efficiency among the tested chambers, which is expected since the electrode separation is the smallest. Even though the Advanced Markus, the Nano Razor and the CC01 chambers have a smaller sensitive volume than the PPC05, their larger electrode spacing results in higher ion recombination. A comparison between the methods used to evaluate for the different chambers is shown in Fig. 6b and Table 4. The assumption of linearity of our measurements can be established, as one can observe in Fig. 3a that for FLASH dose rates all chambers are in the linear regime (R2 > 0.9), except for the PPC05 chamber; therefore, the TVM and Boag’s methods should be applicable. The observed differences of ks values between the different models are within the experimental uncertainties, with the exception of the Nano Razor chamber. For the Nano Razor chamber, these results seem to suggest that initial recombination has the largest effect on the charge collection efficiency while volume recombination is negligible even at the Bragg peak, which is possibly explained by the chamber’s very small volume and electrode spacing. While the increase in initial recombination from the plateau region to the Bragg peak is consistent with the increase of LET at the Bragg peak, which in turn increases same track recombination, this does not explain the observed drop in collection efficiency in VHDR irradiations. Further investigations are warranted in order to verify if other Nano Razor chambers also exhibit this behaviour at high dose rates. Moreover, it would be interesting to test in other irradiation conditions such as with different spot distributions and dose rates, as well as the dependence on the position of the chamber’s stem in relation to the beam axis. Fig. 7 shows the volume recombination coefficient of the Advanced Markus chamber, measured at the entrance plateau of the Bragg peak, as a function of the instantaneous dose rate and Boag’s model [
] for the respective dose rates, demonstrating that the collection efficiency can deviate significantly from this model at the highest dose rates. This model works on the assumption that, upon irradiation, the density of the generated positive ions is uniform across the chamber, which is not the case for single pencil beams. For that reason, Han et al.[
] have proposed a modification to Boag’s theory in which they model the transverse ionised charge spatial distribution in the chamber with a Gaussian distribution to calculate the integral collection efficiency.
Fig. 6(a) Recombination coefficients calculated using the Niatel model. Measurements at the plateau, corresponding to a dose rate of 177 Gy/s, and at the Bragg peak corresponding to 385 Gy/s, are shown for all the probed chambers. (b) ks values obtained with Boag’s model, the TVM and the Niatel model at the plateau region of the Bragg peak. The error bars correspond to statistical uncertainties of 1 σ.
Fig. 7The ion recombination coefficient as a function of the instantaneous dose rate for the Advanced Markus chamber and Boag’s recombination model, corresponding to equation (5).
The PTW microDiamond detector measured a charge of 5.24 nC in CONV (mean dose rate of 4.1 Gy/s) and 5.02 nC in VHDR conditions (mean dose rate of 177 Gy/s) which translates to a 4 % reduction in charge collection efficiency. It is possible that this is a consequence of the dependence with the series resistance and the sensitivity of the commercially available microDiamonds, as it has been shown recently for UHDR electron beams [
]. The Razor diode, on the other hand, presented an even larger reduction; it measured a charge of 17.58 nC in CONV and 10.68 nC in the VHDR regime, which corresponds to a 39 % drop in collected charge.
4. Discussion
Commercially available ionization chambers used in proton VHDR dosimetry need to be corrected for the drop in charge collection efficiency with the dose rate due to ion recombination. Recombination coefficients were evaluated at the entrance of the plateau and at the Bragg peak of a 226.899 MeV proton beam using the Niatel model, the TVM, and Boag’s analytical formula for continuous beams. A general good agreement was obtained between the three techniques with a maximum percentage difference of 2.6 % between Niatel’s model and TVM. Fig. 3a and 3b show that the inverse of the chamber response as a function of the inverse square voltage is linear with a linearity coefficient, R2, higher than 0.9 for all examined detectors except the PPC05, giving us confidence that TVM can be applied. However, in Fig. 4 the curves of the PPC05 chamber appear quite flat with the parameter α equal to 0 for the plateau region and the Bragg peak, and β much smaller than for the other chambers. This may indicate that the chamber is operating in the saturation region. It is worth noting that the PPC05 chamber, with its very small electrode gap, seems especially suited for measurements in VHDRs with recombination coefficients smaller than 0.3 %, an observation also made by Rossomme et al. [
], while the CC01 chamber also presented small recombination coefficients even for the highest dose rates. Surprisingly, we found that the Nano Razor chamber had the highest recombination coefficient (3.6 %) among the probed chambers, even though the sensitive volume is the smallest. However, it is worth mentioning that the very small volume of the chamber might result in a higher susceptibility to dose inhomogeneities and pencil beam positioning, which can adversely increase the uncertainty in the comparison between VHDR and CONV irradiations.
For conventional irradiations our results agree with Yin et al. [
] who found ks to be 1.003 for a mean dose rate of 1.45 Gy/s and 1.006 for 127.58 Gy/s in a DS beam line with the same energy, while we found 1.002–1.008 for 4.1 Gy/s (CONV) and 1.007–1.008 for 177 Gy/s (VHDR) for the Advanced Markus chamber. Similarly, Cunningham et al. [
] measured recombination coefficients smaller than 1.0 % in a FLASH PBS beam with mean dose rates up to 115 Gy/s. It should also be noted that the chamber’s data sheet indicates that for a continuous beam the collection efficiency is ≥ 99.5 % at 187 Gy/s and ≥ 99.0 % at 375 Gy/s, regardless of beam quality. Furthermore, in VHDR conditions our results agree with Yin et al. [
] who found recombination coefficients below 1 % from unity for a variety of parallel plate chambers in a 230 MeV double-scattered proton beam with a maximum mean dose rate of 127.6 Gy/s. The results obtained in this work, therefore, confirm that standard dosimetric equipment can be used in pre-clinical FLASH experiments and clinical trials with adequate accuracy, since most probed chambers in this study have recombination coefficients smaller than 1 %.
5. Conclusion
Accurate and reliable dosimetry is a prerequisite for the implementation and understanding of preclinical FLASH-RT radiobiological experiments and mechanisms. For that reason, it is crucial to accurately assess and report the absorbed dose, the dose rate and the time structure of the irradiation beams. In this study, we evaluated correction factors to be applied due to the reduction of the ion collection efficiency, for both proton CONV and VHDR conditions, for various vented ionization chambers commonly used in radiotherapy. With the exception of the Nano Razor chamber, we find that the probed chambers are suitable for absorbed dose measurements in proton VHDR scanned beams with mean dose rates up to 385 Gy/s, provided that ion recombination coefficients are taken into account.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors acknowledge IBA for the support in using the machine in FLASH mode. This work was partly funded by the European Union’s Horizon 2020 research and innovation program under grant agreement No. 730983 (INSPIRE). This work was partly done in the framework of the project 18HLT04 UHDpulse, which has received funding from the EMPIR program, co-financed by the Participating States and from the European Union’s Horizon 2020 research and innovation program. The simulations for this work were carried out using the access to the HPC resources of TGCC under the allocation 2021-A0100312448 made by GENCI.
References
Dewey D.L.
Boag J.W.
Modification of the oxygen effect when bacteria are given large pulses of radiation.
Fouillade C, Curras-Alonso S, Giuranno L, Quelennec E, Heinrich S, Bonnet-Boissinot S, et al. FLASH irradiation spares lung progenitor cells and limits the incidence of radio-induced senescence. Clin Cancer Res 2020;26:1497–506. https://doi.org/10.1158/1078-0432.CCR-19-1440.
Reduced cognitive deficits after FLASH irradiation of whole mouse brain are associated with less hippocampal dendritic spine loss and neuroinflammation.
Montay-Gruel P, Acharya MM, Jorge PG, Petit B, Petridis IG, Fuchs P, et al. Hypofractionated FLASH-RT as an Effective Treatment against Glioblastoma that Reduces Neurocognitive Side Effects in Mice. Clin Cancer Res 2021;27:775–84. https://doi.org/10.1158/1078-0432.CCR-20-0894.
Feasibility Study of FLASH Radiotherapy for the Treatment of Symptomatic Bone Metastases ClinicalTrials. https://clinicaltrials.gov/ct2/show/NCT04592887.
Rossomme S, Delor A, Lorentini S, Vidal M, Brons S, Jkel O, et al. Three-voltage linear method to determine ion recombination in proton and light-ion beams. Phys Med Biol 2020;65:045015. https://doi.org/10.1088/1361-6560/ab3779.
Toscano S, Souris K, Gomà C, Barragán-Montero A, Puydupin S, Stappen F vander, et al. Impact of machine log-files uncertainties on the quality assurance of proton pencil beam scanning treatment delivery. Phys Med Biol 2019;64:095021. https://doi.org/10.1088/1361-6560/AB120C.
Gomà C, Sterpin E. Monte Carlo calculation of beam quality correction factors in proton beams using PENH. Phys Med Biol 2019;64:185009. https://doi.org/10.1088/1361-6560/AB3B94.
03314-2&id=gr1.jpg){kind=link}
03314-2&id=gr2.jpg){kind=link}
03314-2&id=gr3.jpg){kind=link}
03314-2&id=gr4.jpg){kind=link}
03314-2&id=gr5.jpg){kind=link}
03314-2&id=gr6.jpg){kind=link}
03314-2&id=gr7.jpg){kind=link}