Treatment planning consideration for very high-energy electron FLASH radiotherapy

van de Water S.
Safai S.
Schippers J.M.
Weber D.C.
Lomax A.J.

Towards FLASH proton therapy: the impact of treatment planning and machine characteristics on achievable dose rates.

,

5

Harrington KJ. Ultrahigh Dose-rate Radiotherapy: Next Steps for FLASH-RT. Clin Cancer Res 2019;25(1):3–5.

Google Scholar

]. Due to this significant feature, it has recently attracted extensive interest in many research communities. Many exciting results have been shown in pre-clinical experiments utilizing electron and photon [

6

Schüler E.
Trovati S.
King G.
Lartey F.
Rafat M.
Villegas M.
et al.

Experimental Platform for Ultra-high Dose Rate FLASH Irradiation of Small Animals Using a Clinical Linear Accelerator.

,

7

Spitz D.R.
Buettner G.R.
Petronek M.S.
St-Aubin J.J.
Flynn R.T.
Waldron T.J.
et al.

An integrated physico-chemical approach for explaining the differential impact of FLASH versus conventional dose rate irradiation on cancer and normal tissue responses.

,

8

Montay-Gruel P.
Acharya M.M.
Petersson K.
Alikhani L.
Yakkala C.
Allen B.D.
et al.

Long-term neurocognitive benefits of FLASH radiotherapy driven by reduced reactive oxygen species.

,

9

Loo B.W.
Schuler E.
Lartey F.M.
Rafat M.
King G.J.
Trovati S.
et al.

(P003) Delivery of Ultra-Rapid Flash Radiation Therapy and Demonstration of Normal Tissue Sparing After Abdominal Irradiation of Mice.

,

Favaudon V.
Caplier L.
Monceau V.
Pouzoulet F.
Sayarath M.
Fouillade C.
et al.

Ultrahigh dose-rate FLASH irradiation increases the differential response between normal and tumor tissue in mice.

]. First identified by Hornsey and Bewley in the 1970s, they found a less adverse effect in mouse skin with a higher dose rate irradiation (5000 Gy/min) compared to a lower dose rate of 700 Gy/min [

[11]

Field S.B.
Bewley D.K.

Effects of dose-rate on the radiation response of rat skin.

]. More than three decades later, Favaudon et al. [

[10]

Favaudon V.
Caplier L.
Monceau V.
Pouzoulet F.
Sayarath M.
Fouillade C.
et al.

Ultrahigh dose-rate FLASH irradiation increases the differential response between normal and tumor tissue in mice.

] found that the ultra-high dose rate irradiation (>40 Gy/s) offered a significant decrease of side effects of normal tissue in mice thorax irradiation while maintaining similar tumor control compared to conventional irradiation. Other promising results were also reported in the skin [

[12]

Marie-Catherine V, et al. The advantage of Flash radiotherapy confirmed in mini-pig and cat-cancer patients. Clin Cancer Res 2019;25(1):35–42.

Google Scholar

], head [

8

Montay-Gruel P.
Acharya M.M.
Petersson K.
Alikhani L.
Yakkala C.
Allen B.D.
et al.

Long-term neurocognitive benefits of FLASH radiotherapy driven by reduced reactive oxygen species.

,

13

Montay-Gruel P.
Bouchet A.
Jaccard M.
Patin D.
Serduc R.
Aim W.
et al.

X-rays can trigger the FLASH effect: Ultra-high dose-rate synchrotron light source prevents normal brain injury after whole brain irradiation in mice.

,

14

Simmons D.A.
Lartey F.M.
Schüler E.
Rafat M.
King G.
Kim A.
et al.

Reduced cognitive deficits after FLASH irradiation of whole mouse brain are associated with less hippocampal dendritic spine loss and neuroinflammation.

,

15

Montay-Gruel P.
Petersson K.
Jaccard M.
Boivin G.
Germond J.-F.
Petit B.
et al.

Irradiation in a flash: Unique sparing of memory in mice after whole brain irradiation with dose rates above 100 Gy/s.

], and abdominal tissues [

[9]

Loo B.W.
Schuler E.
Lartey F.M.
Rafat M.
King G.J.
Trovati S.
et al.

(P003) Delivery of Ultra-Rapid Flash Radiation Therapy and Demonstration of Normal Tissue Sparing After Abdominal Irradiation of Mice.

].

Previous FLASH experiments were most performed utilizing the low-energy electron beams (4–20 MeV) [

9

Loo B.W.
Schuler E.
Lartey F.M.
Rafat M.
King G.J.
Trovati S.
et al.

(P003) Delivery of Ultra-Rapid Flash Radiation Therapy and Demonstration of Normal Tissue Sparing After Abdominal Irradiation of Mice.

,

Favaudon V.
Caplier L.
Monceau V.
Pouzoulet F.
Sayarath M.
Fouillade C.
et al.

Ultrahigh dose-rate FLASH irradiation increases the differential response between normal and tumor tissue in mice.

,

12

Marie-Catherine V, et al. The advantage of Flash radiotherapy confirmed in mini-pig and cat-cancer patients. Clin Cancer Res 2019;25(1):35–42.

Google Scholar

,

15

Montay-Gruel P.
Petersson K.
Jaccard M.
Boivin G.
Germond J.-F.
Petit B.
et al.

Irradiation in a flash: Unique sparing of memory in mice after whole brain irradiation with dose rates above 100 Gy/s.

], which do not allow treatment of deep-seated tumors and are always accompanied by the large lateral penumbra. Given the inherent drawbacks of low-energy electrons, there has been increasing interest in combining FLASH with the facilities capable of delivering very high-energy electron (VHEE) beams. After decades of studies into linear colliders, it is achievable to build a compact high-gradient (∼100 MV/m) LINAC making the very high-energy electron accelerator a reality. For very high-energy electrons (50–250 MeV), the penetration depth becomes deeper, and the transverse penumbra shaper, more suitable for deep-seated tumors (>5 cm) [

16

Papiez L.
DesRosiers C.
Moskvin V.

Very high energy electrons (50–250 MeV) and radiation therapy.

,

17

Fuchs T.
Szymanowski H.
Oelfke U.
Glinec Y.
Rechatin C.
Faure J.
et al.

Treatment planning for laser-accelerated very-high energy electrons.

,

18

Åsell M.
Hyödynmaa S.
Söderström S.
Brahme A.

Optimal electron and combined electron and photon therapy in the phase space of complication-free cure.

,

19

Gustafsson A.
Lind B.K.
Brahme A.

A generalized pencil beam algorithm for optimization of radiation therapy.

,

20

DesRosiers C.
Moskvin V.
Bielajew A.F.
Papiez L.

150–250 MeV electron beams in radiation therapy.

]. Furthermore, the intrinsically higher dose compared to photons and protons is more likely to provide a higher dose rate satisfying the FLASH dose rate requirement. Previous studies also demonstrated that the VHEE radiotherapy was capable of intermediate dosimetric advantage between photon VMAT and proton PBS for normal tissue sparing [

Schüler E.
et al.

Very high-energy electron (VHEE) beams in radiation therapy; Treatment plan comparison between VHEE, VMAT, and PPBS.

,

Bazalova‐Carter M.
Qu B.
Palma B.
Hårdemark B.
Hynning E.
Jensen C.
et al.

Treatment planning for radiotherapy with very high-energy electron beams and comparison of VHEE and VMAT plans.

]. Another VHEE study successfully delivered FLASH dose rates at a deep seated prostate tumor [

[23]

Sarti A.
et al.

Deep Seated Tumour Treatments With Electrons of High Energy Delivered at FLASH Rates: The Example of Prostate Cancer.

]. Recently, a promising preliminary design of VHEE linac (a compact C-band system) for FLASH RT has been performed [

[24]

Faillace L.
et al.

Perspectives in linear accelerator for FLASH VHEE: Study of a compact C-band system.

]. At the PITZ, an R&D platform for very high energy electron FLASH radiation therapy and radiation biology is being prepared (FLASHlab@PITZ) [

[25]

Stephan F.
et al.

FLASH l a b @PITZ: New R&D platform with unique capabilities for electron FLASH and VHEE radiation therapy and radiation biology under preparation at PITZ.

]. Clinical FLASH irradiations also come with dosimetric challenges. In another study of UHDR pulsed electron beam, a first characterization of 6 real-time point scintillating dosimeters with 5 phosphors was performed. Overall, the VHEE beams were capable of great expectation to bring FLASH to the clinic [

[26]

Vanreusel V.
et al.

Point scintillator dosimetry in ultra-high dose rate electron & #x201c;FLASH” radiation therapy: A first characterization.

]. Overall, the VHEE beams were capable of great expectation to bring FLASH to the clinic [

23

Sarti A.
et al.

Deep Seated Tumour Treatments With Electrons of High Energy Delivered at FLASH Rates: The Example of Prostate Cancer.

,

24

Faillace L.
et al.

Perspectives in linear accelerator for FLASH VHEE: Study of a compact C-band system.

,

25

Stephan F.
et al.

FLASH l a b @PITZ: New R&D platform with unique capabilities for electron FLASH and VHEE radiation therapy and radiation biology under preparation at PITZ.

,

26

Vanreusel V.
et al.

Point scintillator dosimetry in ultra-high dose rate electron & #x201c;FLASH” radiation therapy: A first characterization.

,

27

Ronga M.G.
Cavallone M.
Patriarca A.
Leite A.M.
Loap P.
Favaudon V.
et al.

Back to the Future: Very High-Energy Electrons (VHEEs) and Their Potential Application in Radiation Therapy.

,

28

Poppinga D.
Kranzer R.
Farabolini W.
Gilardi A.
Corsini R.
Wyrwoll V.
et al.

VHEE beam dosimetry at CERN Linear Electron Accelerator for Research under ultra-high dose rate conditions.

].

In our study, the primary purpose is to investigate the potential FLASH-related performance (plan quality, dose rate, and delivery time) for VHEE beams (PBS delivery mode). Our study comprises three steps (Fig. 1). First, irradiated by different beam energies, the dose-influence matrix was calculated using a scanning beam for two clinical cases (head: 80 MeV, 100 MeV, and 120 MeV, lung: 100 MeV, 120 MeV, and 140 MeV). Second, the VHEE treatment plans for two patient targets were obtained via a PTV-based optimization algorithm (derived from the proton PBS plan). Third, the quantitative analyses were made in terms of treatment plan quality and FLASH-related parameters (DVH, nCI, DADR, and total duration of treatment).

Figure thumbnail gr1 — Fig. 1The schematic workflow in our study.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

Methods and materials

Dose-influence matrix calculation and PTV-based optimization method

We used the Monte-Carlo method (Geant4) to calculate the dose-influence matrix (K_ij) for VHEE plan optimization in two patient targets (head and lung cases). In Eq. (1), Di represents the total dose in the voxel i. Kij represents the dose contribution per unit weight of the voxel i at the jth spot. The non-negative quantity wj2 refers to the weight at the jth spot. The constructed dose-influence matrix contains the information of VHEE beams irradiated at different scanning spots under different fields (Fig. 2). The PTV-based optimization method utilizing a standard quadratic objective function was adopted to obtain the optimal spot weights [

29

Zhang X.
Liu H.
Wang X.
Dong L.
Wu Q.
Mohan R.

Speed and convergence properties of gradient algorithms for optimization of IMRT.

,

30

Oelfke U.
Bortfeld T.

Inverse planning for photon and proton beams.

,

31

Wu Q.W.
Mohan R.

Algorithms and functionality of an intensity modulated radiotherapy optimization system.

]. Each optimized spot weight (electrons) determines the corresponding beam duration. The overall treatment time is the summation of beam duration of different fields and scanning time. Details are in Eq. (2). Here p refers to the penalty weight. D0 corresponds to the prescribed dose of the target or constraint in OARs. H is the Heavyside function (Eq. (3)), and the voxel dose (Di) was from the Eq. (1). Incorporating the planning target volume (PTV) definition into the optimization, this algorithm considered the setup uncertainties by simply expanding the CTV outward with a 3 mm safety margin (PTV). Unlike proton plan optimization, there was no need to consider the range uncertainties in VHEE plan optimization. The optimal spot weights were obtained by minimizing the objective function via a large-scale non-linear solver IPOPT developed in MATLAB interface [

32

Liu W.
Zhang X.
Li Y.
Mohan R.

Robust optimization of intensity modulated proton therapy.

,

33

Wachter A.
Biegler L.T.

On the implementation of an interior-point filter line-search algorithm for large-scale nonlinear programming.

,

34

Wachter A.
Biegler L.T.

Line search filter methods for nonlinear programming: Motivation and global convergence.

,

35

Wachter A.
Biegler L.T.

Line search filter methods for nonlinear programming: Local convergence.

]. After optimization, the created treatment plans for two patients were made sure to reach the standard of the clinically acceptable plan.

D_{i} = \sum_{j} K_{ij} w_{j}^{2}

(1)

F_{ptv} (w_{j}) = \sum_{i ∊ P T V} p_{ptv} {(D_{i} - D_{0, p t v})}^{2} + \sum_{i ∊ P T V} p_{OARs} H (D_{i} - D_{0, O A R s}) {(D_{i} - D_{0, O A R s})}^{2}

(2)

H (t) = \{\begin{matrix} 0, t \leq 0 \\ 1, t \geq 0 \end{matrix}

(3)

Figure thumbnail gr2 — Fig. 2Scanning spots at 50°, 120°, and 330° for head case optimized for VHEE radiation therapy. The scanning spots in each gantry angle spread all over the region of PTV + 6 mm with a spot spacing of 3 mm (only shown head case).
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

Patient cases

Six treatment plans (each patient irradiated by three VHEE beam energies) were created for two patient cases (head and lung). One tumor (head case) locates about 80 mm from the skin surface, and another tumor is about 170 mm from the skin. The region surrounding PTV was always accompanied by the high delivered dose (similar to PTV). It was crucial for this region to receive FLASH dose rate achieving normal tissue sparing. Therefore, the OAR was assigned with a width of 2 cm (head) and 1 cm (lung) extending from the edge of the PTV (Fig. 3, Fig. 4). The head case was irradiated with VHEE beams at 80 MeV, 100 MeV, and 120 MeV, each with 16 beam angles (330° to 120° with 10° interval). Another lung tumor was irradiated at 100 MeV, 120 MeV, and 140 MeV, each with 11 beam angles (130° to 230° with 10° interval). For head and lung cases, the total number of scanning spots (VHEE beams) was 4655 and 1591, respectively. The specific spot patterns at three beam angles were shown in Fig. 2. Both patient cases were irradiated at a constant beam intensity under each scanning spot. The size of the VHEE beam was assumed of 0.3–5 mm (80–140 MeV) defined by the full width at half maximum (FWHM) [

[22]

Bazalova‐Carter M.
Qu B.
Palma B.
Hårdemark B.
Hynning E.
Jensen C.
et al.

Treatment planning for radiotherapy with very high-energy electron beams and comparison of VHEE and VMAT plans.

]. The dose grid resolution was 2.73 × 2.73 × 3.00 mm³ (head) and 2.34 × 2.34 × 1.50 mm³ (lung).

Figure thumbnail gr3 — Fig. 3Structure volume (PTV, OAR) and the results of dose distribution, DVH, and plan quality-related parameters (V_42Gy, V_45Gy, and nCI) for 80 MeV, 100 MeV, and 120 MeV, for the head case.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

Figure thumbnail gr4 — Fig. 4Structure volume (PTV, OAR) and the results of dose distribution, DVH, and plan quality-related parameters (V_42Gy, V_45Gy, and nCI) for 100 MeV, 120 MeV, and 140 MeV, for lung case. The plan quality parameter related to OAR (V_25Gy) is involved in the results.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

Quality comparison and FLASH-related analyses

The plan qualities of two patient cases were compared in terms of dose-volume histogram (DVH), nCI (CI/target coverage), and individual OAR dose statistics. It should be emphasized that the same optimization parameters (penalty weight, prescription dose, and OAR constraint) were adopted for different patient cases (head, lung).

Our study mainly focused on two potential parameters related to the FLASH effect, which was detailed below. All FLASH-related results were analyzed at a fraction dose of 10 Gy, and the beam intensity remain constant at each scanning spot.

(a)
Dose Rate

Previous studies have shown that dose rate plays an essential role in achieving the FLASH effect. Here, we considered the dose rate higher than 40 Gy/s capable of FLASH benefit. There is still no general agreement about the most useful definition of 'Dose Rate' in PBS mode. The 3D dose-averaged dose rate (DADR) will be a reasonable choice to define the dose rate under the scanning beam (spot) [

van de Water S.
Safai S.
Schippers J.M.
Weber D.C.
Lomax A.J.

Towards FLASH proton therapy: the impact of treatment planning and machine characteristics on achievable dose rates.

,

36

Schwarz M.
Traneus E.
Safai S.
Kolano A.
van de Water S.

Treatment planning for Flash radiotherapy: General aspects and applications to proton beams.

]. The dose-averaged dose rate (Gy/s) was detailed in Eq. (4). For DADR calculation, beamlet sizes of 5 mm FWHM have been proposed in our study. For head case, there needs 4655 spots for 16 irradiation beams. Another lung tumor was irradiated with 1591 spots for 11 irradiation beams. Each spot irradiation duration influences the value of DADR (without considering the dead time). However, the single beam duration is determined by the corresponding spot weight solved by the objective function.

{DADR}_{i} = \sum_{j = 1}^{n} \frac{{(d}_{ij} w_{j}) {(d}_{ij} {BI}_{j})}{\sum_{j = 1}^{n} {(d}_{ij} w_{j})}

(4)

where:

D A D R d o s e - a v e r a g e d d o s e r a t e (G y / s);

i s i n g l e v o x e l;

j s c a n n i n g s p o t;

w s p o t w e i g h t (e l e c t r o n s);

d d o s e - i n f l u e n c e m a t r i x (G y / e l e c t r o n);

$B I b e a m i n t e n s i t y (e l e c t r o n s / s)$ ;

(b)
Overall irradiation time (beam-on time + scanning time)

Both the spot irradiation times and beam irradiation times were calculated. The beam irradiation time is the sum of all individual spot irradiation times, whereas the total beam time also includes time needed for scanning. The overall irradiation time of an entire plan consists of the sum of irradiation times of the individual beams (head case: 4655 spots, 16 beams, lung case: 1591 spots, 11 beams). The scanning speed of spot-to-spot and line switching is 40 mm/ms and 10 mm/ms. The overall irradiation time it has been reported that the time scale ∼500 ms and < 200 ms were potentially related to the FLASH effect [

van de Water S.
Safai S.
Schippers J.M.
Weber D.C.
Lomax A.J.

Towards FLASH proton therapy: the impact of treatment planning and machine characteristics on achievable dose rates.

,

Favaudon V.
Caplier L.
Monceau V.
Pouzoulet F.
Sayarath M.
Fouillade C.
et al.

Ultrahigh dose-rate FLASH irradiation increases the differential response between normal and tumor tissue in mice.

,

37

Bourhis J.
et al.

Clinical translation of FLASH radiotherapy: Why and how?.

,

38

Vozenin M.C.
Hendry J.H.
Limoli C.L.

Biological Benefits of Ultra-high Dose Rate FLASH Radiotherapy: Sleeping Beauty Awoken.

]. In our study, we calculated all the irradiation times at a fraction dose of 10 Gy and with a constant beam intensity at each scanning spot.

Results

Plan quality comparison

Head case

After optimization, the plan quality-related parameters, including dose map, DVH, V_42Gy/V_45Gy, and nCI, are all shown in Fig. 3. The dose maps (Fig. 3, middle column) involve three beam energies (80 MeV, 100 MeV and 120 MeV). The maximum dose of 120 MeV is higher than plans utilizing 80 MeV and 100 MeV. Moreover, the proportion of the volume of the dose > 45 Gy in the PTV is lower than the other two energies. From the DVH, the curve of PTV of 120 MeV was slightly worse than 80 MeV and 100 MeV (Fig. 3, DVH). Besides these, the DVH of the three beam energies in the head case shows no big difference. The nCI of the PTV is 1.09 (80 MeV), 1.10 (100 MeV) and 1.14 (120 MeV), and the V_42Gy/V_45Gy refers to 0.99/0.98 (80 MeV), 0.99/0.98 (100 MeV), 0.99/0.97 (120 MeV). On the whole, all VHEE plans (80 MeV, 100 MeV, and 120 MeV) of the head case satisfy the requirements of a clinical treatment plan.

Lung case

Similar to the head case, Fig. 4 (middle column) shows three dose distribution maps with different beam energies (100 MeV, 120 MeV, 140 MeV). Unlike the head case, the treatment plan with higher energy offers a better plan quality than lower energy. To be specific, the nCI is 1.17 (100 MeV), 1.16 (120 MeV) and 1.16 (140 MeV). The V_42Gy/V_45Gy in the PTV refers to 0.99/0.95 (100 MeV), 0.99/0.96 (120 MeV), 0.99/0.96 (140 MeV). In the lung case, we added the parameter of V_25Gy to evaluate the volume of OAR receiving dose < 25 Gy (Fig. 4, table) given the obvious difference in the DVH of OAR (Fig. 4, DVH, OAR). The V_25Gy of OAR of 140 MeV is 0.72, lower than 0.78 of 120 MeV and 0.83 of 100 MeV. All VHEE plans of the lung case still satisfy the clinical treatment plan requirement.

FLASH-related comparison

Head case

Using the definition of dose-averaged dose rate (DADR), the dose rate maps and DRVHs have been shown in Fig. 5. Also shown in the picture is the irradiation time varying with the number of scanning spots. It requires a beam intensity of ∼2.5e + 11 electrons/s to allow > 90 % volume of PTV receiving a DADR > 40 Gy/s, no matter the energy of 80 MeV, 100 MeV and 120 MeV. At this beam intensity, the dose rate of 120 MeV in each voxel of the PTV is at an average of ∼21 % and ∼44 % higher than that of 100 MeV and 80 MeV, respectively. In each voxel of OAR, this number is ∼20 % and ∼42 %. Also, it can be found in the DRVH that the dose rate of higher energy is always higher than that in the lower energy for both OAR and PTV (Fig. 5, DRVH). From the dose rate map (beam intensity: 2.5 × 10¹¹ electrons/s), there are always voxels in normal tissue (OAR) receiving the dose rate higher than that in the target volume (PTV). From DRVH (beam intensity: 2.5 × 10¹¹ electrons/s), the volume receiving the high dose rate (DADR) in the OAR is bigger than that in the PTV (Fig. 5, DRVH) and this makes no difference between energies.

Figure thumbnail gr5 — Fig. 5The head case. Dose rate (DADR) distributions for three VHEE beam energies (80 MeV, 100 MeV, and 120 MeV) are shown at the top. The dose rate volume histogram (DRVH) and the irradiation time varying between scanning spots are shown at the bottom. All the results are calculated at the beam intensity of 2.5 × 10¹¹ electrons/s.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

The irradiation time in Fig. 5 is also calculated at the beam intensity of 2.5 × 10¹¹ electrons/s (constant beam intensity) and a fraction dose of 10 Gy. The total irradiation time is 5258.75 ms (80 MeV), 5149.75 ms (100 MeV) and 4976.75 ms (120 MeV), including scanning time 872.75 ms. The averaged irradiation time (only considering the beam-on time) per scanning spot is ∼1.13 ms (80 MeV), ∼1.11 ms (100 MeV), and ∼1.07 ms (120 MeV).

Lung case

For the lung case, the FLASH-related parameters (dose rate and irradiation time/beam duration) are shown in Fig. 6. Similar results are seen for the lung case. The dose rate (DADR) in the region surrounding PTV is comparable to that in the PTV, somewhere even higher. From the DRVH, the high energy demonstrates a higher dose rate than low energy for both OAR and PTV. Unlike the head case, the beam intensity needs to be ∼9.4e + 11 electrons/s to let the > 90 % volume of PTV receive the dose rate higher than 40 Gy/s. At this beam intensity, the dose rate of 140 MeV in each voxel of the PTV is at an average of ∼24 % and ∼43 % higher than that of 120 MeV and 100 MeV, respectively. In OAR, these number refer to ∼21 % and ∼39 %.

Figure thumbnail gr6 — Fig. 6The lung case. The top shows dose rate (DADR) distributions for three VHEE beam energies (100 MeV, 120 MeV, and 140 MeV). The dose rate volume histogram (DRVH) and the irradiation time varying between scanning spots are shown at the bottom. All the results are calculated at the beam intensity of 9.375 × 10¹¹ electrons/s.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

In Fig. 6, the dose rate map, DRVH, and irradiation time are all calculated at the beam intensity of 9.375 × 10¹¹ electrons/s. The total irradiation time is 1034.25 ms (100 MeV), 981.55 ms (120 MeV), and 928.15 ms (140 MeV), including scanning time 397.75 ms. For each spot, the averaged irradiation time is ∼0.65 ms (100 MeV), ∼0.61 ms (120 MeV), and ∼0.58 ms (140 MeV).

Discussion

This study systematically investigates the potential FLASH-related parameters for VHEE treatment plans using head and lung irradiation as the paradigm. Furthermore, the dosimetry performances (V_42Gy/V_45Gy, DVH, nCI) and physical parameters related to the FLASH effect (dose rate (DADR), irradiation time) were all quantitative. They successfully established a benchmark reference for VHEE FLASH dose rate performance (e.g., 40 Gy/s), even the FLASH-related threshold (e.g., 40 Gy/s) changes in the future. To better understand the impact of beam energy on the dose rate (or irradiation time), we involved three beam energies in each patient case and made a quantitative comparison for their FLASH-related parameters. It should be clarified that the different delivery time is only related with the different interactions of electrons with the tissue. According several observations, the flash effect can occur just if the delivered dose is beyond a certain threshold. To provide a definitive answer of this threshold is beyond our capacity, especially when the underlying processes associated with FLASH remain unclear. In reality, it is currently unclear whether the instantaneous dose rate within a given pulse or the average dose rate over several pulses within a spot is most relevant (ie, the importance of “dead-time” during FLASH delivery is still uncertain). As long as this uncertainty persists, there is value in exploring various scenarios: in this work, we make the assumption that the dose rate/spot is the key parameter in determining FLASH sparing. Until very recently, the experimental evidence shows that all the following constraints may be respected to trigger the FLASH benefits [

[39]

Vozenin M.C.
Bourhis J.
Durante M.

Towards clinical translation of FLASH radiotherapy.

], a) ADR < 40 Gy/s (but better 100), b) Dose-per-pulse > 1 Gy, c) irradiation time > 200 ms (but better 100), d) Total dose > 8 Gy. Flash effect is important not for the target tissue (higher dose value) but for the OARs. The experimental evidence has shown that the total dose > 8 Gy may be respected to trigger the FLASH benefits. From the DVH (Fig. 3, Fig. 4), the volume of OARs receiving dose higher than 8 Gy were < 25 % (head) and < 50 % (lung) when the fraction dose of 10 Gy was considered. As the fraction dose increased to 20 Gy, these volumes were changed to ∼80 % (head) and ∼90 %. The most volume of the defined OARs receiving dose > 8 Gy when the fraction dose reached 20 Gy. In our study, we utilized a fraction dose of 10 Gy for irradiation time (beam duration) calculation. The irradiation time can be simply obtained by multiplying a dose threshold modify factor even the dose threshold triggering FLASH benefits has changed. It should be noted that the dose threshold of triggering the FLASH benefits will not affect the dose rate distribution (highly depends on VHEE beam intensity).

Plan quality

In this paper, the beam angles were selected by an experienced clinical physicist and referred to the standard IMRT plan and the VHEE plan in reference [

Schüler E.
et al.

Very high-energy electron (VHEE) beams in radiation therapy; Treatment plan comparison between VHEE, VMAT, and PPBS.

,

Bazalova‐Carter M.
Qu B.
Palma B.
Hårdemark B.
Hynning E.
Jensen C.
et al.

Treatment planning for radiotherapy with very high-energy electron beams and comparison of VHEE and VMAT plans.

]. In the reference [

Schüler E.
et al.

Very high-energy electron (VHEE) beams in radiation therapy; Treatment plan comparison between VHEE, VMAT, and PPBS.

,

Bazalova‐Carter M.
Qu B.
Palma B.
Hårdemark B.
Hynning E.
Jensen C.
et al.

Treatment planning for radiotherapy with very high-energy electron beams and comparison of VHEE and VMAT plans.

], the head case was irradiated by 13, 17 and 36 beams and the lung case was irradiated by 16 beams. From the DVHs and nCIs, all VHEE plans for two patient targets meet the requirement of a clinically acceptable plan. Unlike the lung case, a higher energy plan of the head case has not showed with a better plan quality than lower energy. From Fig. 3, the DVH of OAR of high energy is quite comparable to the low energy, and the DVH curve of PTV of 120 MeV performs even worse than 100 MeV and 80 MeV. The same condition also can be found in nCI (Fig. 3, table). The result in our study seems not consistent with previous studies of higher energy capable of better treatment plan quality [

Schüler E.
et al.

Very high-energy electron (VHEE) beams in radiation therapy; Treatment plan comparison between VHEE, VMAT, and PPBS.

,

Bazalova‐Carter M.
Qu B.
Palma B.
Hårdemark B.
Hynning E.
Jensen C.
et al.

Treatment planning for radiotherapy with very high-energy electron beams and comparison of VHEE and VMAT plans.

]. Compared to the lung case, the relatively shallower target may be the leading cause. It is also the main reason why the highest beam energy of the head case was selected to be 120 MeV, not being similar to the lung case of 140 MeV. It also suggests that there is always a fitting energy interval for a given target depth to achieve the optimal treatment plan for the VHEE beams. Obviously, the optimal energy interval for the head case in our study needs to be lower than 120 MeV.

FLASH dose rate

The pencil beam scanning strategy has been widely deployed for intensity-modulated radiation therapy. It offers several advantages, such as better dose conformality and higher beam transport efficiency (higher instantaneous dose rate in a single spot). This naturally lends itself well to achieving FLASH radiation therapy. However, the definition of dose rate in this delivery mode (PBS mode) for FLASH consideration is still controversial. The definition (DADR) proposed in a previous study for PPBS dose rate calculation [

[4]

van de Water S.
Safai S.
Schippers J.M.
Weber D.C.
Lomax A.J.

Towards FLASH proton therapy: the impact of treatment planning and machine characteristics on achievable dose rates.

] is selected in our work. It is designed specifically for the proton pencil beam scanning (PPBS) strategy. The DADR averages the dose rate in a voxel over all of the scanning spots while being weighted by the dose contribution delivered at each scanning spot. The DADR defined in this manuscript has been adopted by several other studies [

van de Water S.
Safai S.
Schippers J.M.
Weber D.C.
Lomax A.J.

Towards FLASH proton therapy: the impact of treatment planning and machine characteristics on achievable dose rates.

,

40

Gao H.
Lin B.
Lin Y.
Fu S.
Langen K.
Liu T.
et al.

Simultaneous dose and dose rate optimization (SDDRO) for FLASH proton therapy.

,

41

Lin Y.
Lin B.
Fu S.
Folkerts M.M.
Abel E.
Bradley J.
et al.

SDDRO-joint: simultaneous dose and dose rate optimization with the joint use of transmission beams and Bragg peaks for FLASH proton therapy.

,

42

Kang M.
Wei S.
Choi J.I.
Simone C.B.
Lin H.

Quantitative Assessment of 3D Dose Rate for Proton Pencil Beam Scanning FLASH Radiotherapy and Its Application for Lung Hypofractionation Treatment Planning.

] although its validity may be somewhat disputable (e.g., ignoring switching time). One essential question is whether the entire time between the first and last dose deposition should be taken into account. To provide a definitive answer is beyond our capacity, especially when the underlying processes associated with FLASH remain unclear for the time being (e.g., what is the time scale oxygen depletion persists?). One study only considered instantaneous dose rates during DADR calculations, ignoring dead times between spots. Another two slightly different definitions of dose rate for FLASH consideration are discussed below. One definition using an arbitrary dose contribution threshold of 0.1 cGy to exclude the spots with low dose contribution [

[43]

van Marlen P.
Dahele M.
Folkerts M.
Abel E.
Slotman B.J.
Verbakel W.F.A.R.

Bringing FLASH to the Clinic: Treatment Planning Considerations for Ultrahigh Dose-Rate Proton Beams.

], the biological meaning of which has not been currently demonstrated. Otherwise, the value of this threshold is difficult to define. The second definition averages the dose rate between spots by applying a correction factor [

[44]

Zhang G.
Wang J.
Wang Y.
Peng H.

Proton FLASH: passive scattering or pencil beam scanning?.

], which yields a lower dose rate compared to the instantaneous counterpart. In our opinion, even the definition of DADR is not a fully rigid metric for characterizing FLASH dose rate, no better alternative exists.

Different from the DVH and dose distribution, the DRVH or dose rate distribution performs that high energy always offers a higher dose rate (DADR) for both PTV and OAR. FLASH or ultrahigh dose rate irradiation has shown a sparing effect in normal tissue (OAR). This leads us to provide a direct comparison (Fig. 5 and Fig. 6) between normal tissue (OAR) and tumor (PTV). In OAR, there is always a volume receiving dose rate higher than PTV, no matter the energy of the VHEE beam. On the other hand, there also exists a region of OAR where the DADR is lower than the FLASH dose rate (40 Gy/s). For better analyzing the low dose rate region (<40 Gy/s) in the OAR, we divided the existing volumes of OAR into three (head: R1, R2, R3) and two (lung: R1, R2) smaller regions (details in Fig. 7). The DVH and DRVH corresponding to R1, R2, and R3 are shown in Fig. 7.

Figure thumbnail gr7 — Fig. 7The dose rate and dose-volume histograms for smaller regions in the OAR, for the head case and lung case. In head case, R1 is 3 mm extending from the edge of the PTV. R2 is 5 mm extending from the edge of R1. R3 is 12 mm extending from the edge of R2. For the lung case, R1 is 3 mm extending from the edge of the PTV. R2 is 7 mm extending from the edge of R1. The OARs for the head and lung are 20 mm and 10 mm surrounding the PTV.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

DADR of R1 (3 mm region outside the PTV) is quite comparable to that of PTV (Fig. 7, left). More importantly, this region (R1) is usually delivered with the highest deposited dose outside the PTV (>80 % volume receives dose between 40 Gy and 45 Gy). Therefore, given the high receiving dose in R1 (comparable to PTV), there also needs to be delivered with high dose rate (>40 Gy/s) for this region to achieve normal tissue sparing. However, for the region farther away from PTV (head: R3, lung: R2), ∼ 60 % (lung) and ∼90 % (head) volume receives dose lower than 35 Gy, and ∼40 % (lung) and ∼50 % (head) volume receives DADR lower than 40 Gy/s. Concluded from Fig. 3 and Fig. 4, the volume receiving a low deposited dose is usually accompanied by a low dose rate. It may be beneficial for normal tissue with a low dose rate (<FLASH dose rate requirement) to reduce the late adverse reaction. The minimum beam intensity needs to be at 2.5 × 10¹¹ electrons/s (head) and 9.375 × 10¹¹ electrons/s (lung) for allowing > 90 % volume of PTV receiving DADR higher than 40 Gy/s. However, it is noted that no matter how high the beam intensity is, there is always a volume in OAR receiving DADR lower than 40 Gy/s.

In the presented analysis, OARs are defined as regions surrounding the PTV. This approach is oversimplified for a proper evaluation of the treatment plan quality. We therefore added the dose and dose rate analysis in real organ at risks (left lung, right lung, and spinal cord), for the lung case. From the DVH and DRVH (Fig. 8), the VHEE beams of higher energy always present the higher dose and dose rate distributions in all the real organ at risks (Fig. 8, DVH and DRVH). In DVH, the V_20Gy/ V_30Gy is much lower than 30 %/20 %, for both left and right lung. The maximum dose in spinal cord is less than 33 Gy (<45 Gy). The dose of all the OARs satisfy the clinical required plan, whatever the beam energy is. From the DRVH, the ratio of volume receiving the FLASH dose rate (>40 Gy/s) in all the OARs is less than 20 %, for all three beam energies. Although the insufficient volume of OAR receiving FLASH dose rate (Fig. 8, DRVH), the low received dose (Fig. 8, DVH) itself does not cause more damage to normal tissue.

Figure thumbnail gr8 — Fig. 8The dose rate and dose-volume histograms for real organs at risk (left lung, right lung, and spinal cord), for the lung case.
View Large Image
Figure Viewer
Download Hi-res image
Download (PPT)

FLASH time scale

As shown in the Fig. 5, Fig. 6, the total irradiation time/beam-on time varied between 4100 and 4400 ms in head case (lung case: 620–740 ms), not including the dead time (e.g., switching time between adjacent spots). The scanning time is 872.75 ms (Head) and 298.75 ms (Lung). As utilize 4655 scanning spots in the head case (lung: 1591), the total irradiation time is ∼6 times longer than in the lung case. It is questionable whether the total irradiation time of head and lung cases in our study is still too long for achieving the FLASH effect. The answer may be twofold. The upper limits of extraction beam intensity of most VHEE facilities have not yet been reached (only 2.5 × 10¹¹ electrons/s and 9.375 × 10¹¹ electrons/s). Deploying a higher extraction beam intensity can reduce the total irradiation time to a 200–500 ms scale [

van de Water S.
Safai S.
Schippers J.M.
Weber D.C.
Lomax A.J.

Towards FLASH proton therapy: the impact of treatment planning and machine characteristics on achievable dose rates.

,