Radiobiological Outcomes, Microdosimetric Evaluations and Monte Carlo Predictions in Eye Proton Therapy

Round 1

Reviewer 1 Report

This paper reports the characteristics under proton beam irradiation using two types of cells and two types of detectors, and calculation of radiobiological effects. The explanation of the content is difficult to understand, and there is no novelty. There are few discussions and few comparisons using citations. The conclusions are similar to the two papers Physica Medica 58 (2019) 72–80 and Phys. Med. Biol. 65 (2020) 245018.

Reject is recommended.

 

Below are comments.

 

Figure 1

I think Figure 1 is a quote from Physica Medica 58 (2019) 72–80, but it's confusing.

There is too much information. The calculated black line is different from the measured value in SOBP. Is this okay? LET of primary and secondary particles is described, but there is no explanation in the text. In addition, although it is described as "reported", there is no quotation.

There is no description of what LET-dose indicates. I don't know (red circles), it's hard to see.

 

Line: 176 What is RBEmu?

Line: 219 The text is strange

Line: 217 I don't know where the results and discussions of the three merit methods are.

 

Line: 217- 232 ”ocular tissue would lead to a loss of function of the entire organ.” This sentence is not a discussion or results. Should be move to the experimental section.

 

Tables 1 and 2 differ only in cell type, so please summarize them. It's hard to understand.

 

 

Line: 246 . “The capability of the two microdosimeters to describe the RBE variation along the proton penetration depth was then investigated.” is the same sentence as the cited reference Phys. Med. Biol. 65 (2020) 245018. In this paper, only two LET variations are found, the entrance and the middle. That's why the novelty is low.

 

The resolution of FIG. 2 is low and cannot be read.

Line: 264 From what data did you calculate RBE 1.23 and 1.12?

Line: 265 Discussion on Table 3 There is no comparison or discussion.

I'm not sure about the points to discuss in Figure 2.

Figure 3 seems to be the important data in this paper, but it seems unlikely that the experimental results and the model of cell irradiation effects show good agreement.

 

Tables 4 and 5 summarize only the differences in cells.

 

I couldn't understand the values in table 4 and 5. It seems to indicate the degree of agreement between the experimental results and the comparison of various models, but please show the raw data compared. If Fig. 3 is a comparison, it is unlikely that the experimental results are reproduced, as written in the conclusion.

Please unify the notation of RBE10.

Author Response

 

 

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper reported the comparison between the experimental measurement of cell response of radiation and the predictions of 3 models. The results show good agreement between the experimental data and the predictions (except for one case).
The main concern is that: 
The cell response at only one position in the beamline (position not clear) was measured. More data would make it better to prove, as claimed in the goal of the paper, either the RBE VARIATIONS, (general) agreement between model prediction and experimental data.
Please find detailed comments in the .pdf file. 

Comments for author File: Comments.pdf

Author Response

We would like to thank the Reviewer number 2 for his/her valuable concerns and suggestions.

We implement our work basing on reviewer’s suggestion that are summarized in the following point by point respons:

 

Comments and Response:

 

This paper reported the comparison between the experimental measurement of cell response of radiation and the predictions of 3 models. 

The results show good agreement between the experimental data and the predictions (except for one case).

 

The main concern is that: 

The cell response at only one position in the beamline (position not clear) was measured. 

More data would make it better to prove, as claimed in the goal of the paper, either the RBE VARIATIONS, (general) agreement between model prediction and experimental data.

 

R1.

We thank Reviewer #2 for his/her constructive inputs. Moreover, we agree with his/her comment and concerns about the fact that cells were exposed solely at one position along the clinical proton SOBP (precisely, at  a depth of 24 mm) .The current study pinpoints a clinical and critical aspect regarding the  treatment of neoplasms such as the uveal melanoma by proton therapy. Hence, the correct evaluation of the damage induced by ionizing radiation during RT treatment ,also thanks to Monte Carlo-based simulations, represents a valuable and potent tool to foresee the impact of ionizing radiation on cancer tissues. On the other side, it’s equally important to evaluate biological damage induced by an erroneous dose distribution during a proton therapy treatment plan that may have detrimental effects on normal tissue, for such reason the effects on the normal retinal pigment epithelial ARPE-19 cell line were investigated and compared to the ones induced in the uveal melanoma 92.1 cell line. Moreover, as highlighted by recent work by Anfuso et al (2020) and Tosi et al 2021 (https://doi.org/10.1186/s13046-021-01947-1), the uveal melanoma microenvironment is thought to play a decisive role in determining the metatastatic potential of this neoplasm, which very often develops adjacent to the retina and has a high infiltrating capacity. Therefore, the uveal tumour microenvironment is a complex and intricate milieu possibly comprising normal retinal cells.  Irradiating uveal melanoma cancer cells at the mid-SOBP position, as performed in our work, often therefore means irradiating normal retinal cells as well, the radioresponse of which is currently little known.  Obviously, future work will have to address the impact of normal retinal tissue exposure at the distal proton SOBP. Nevertheless, we believe that this is an important starting point to evaluate in vitro the agreement between microdosimetry-based calculations and biological effects in a realistic therapeutic scenario.   

 

Line 4: 10% increase is not 'slightly

R2.

The abstract was modified and implemented

 

Line 12-13: one position is not about variations

 

R3.

The abstract was modified and implemented

 

Line 57: Please modify this: In this paper, only 1 position, which is so-called mid SOBP, is estimated.Usually at least 2 spots are needed to prove some variation.

 

R4.

Text was modified (new line 54) in “The main aim  of this work was to study the RBE values at the clinical SOBP of the CATANA proton therapy facility.”

 

Line 64-65: This damage usually happens at the distal edge of the SOBP (this is also mentioned at the beginning of page 3), where the LET is much higher than that in the 'mid of SOBP'. Hence, the damage analyzed by this work may not directly relate to the 'damage induced by proton therapy in a typical OAR'. Please modify this.

 

R5.

In order to compare the effects induced by proton therapy ,in a non-tumorigenic  cell line  the human normal retinal pigment epithelial ARPE-19 cell line was  also irradiated. The cell line ARPE-19 was chosen as our control for two main reasons: the first, because normal cells can easily recover from damage induced by ionizing radiation thanks to a more efficient damage response machinery in respect to the ones activated in cancer cells; second, because it’s useful to evaluate the detrimental effects caused by even minimal errors of dose deposition in tissues surrounding the tumor during a proton therapy session. Finally, as also mentioned above, the tumour microenvironment is indeed a mixture of cancer and normal cells, therefore, understanding the radiobiological behaviour at mid-SOBP of normal retinal cells may provide the basis for further characterization of the metastatic potential of uveal cancer cells following proton therapy. 

 

According these considerations we modified and implemented the text (from line 60 to 72)

 

Line 75-76-77: Only one position is analyzed by this work. More data is needed to claim the comparison between models and experimental data.

 

R6.

The sentence was modified (new line 84) adding "at a depth of 24 mm, the mid-position of the spread-out Bragg peak"

 

Figure 1: Please add legend. Please also mark where the positions in this work are analyzed (detectors, cells etc.)

 

R7.

The Figure was improved accordingly with the suggestion

 

Line 113: Please give some quantitative description, e.g. x mm shown in the figure 1

 

R8.

Sentence was improved (see new lines 84 and 111) adding the sentence 24 mm, the mid-position of the spread-out Bragg

 

Line 155: Can you please provide the detailed methods how you make the fitting?

As LQ equation is a numeric model to fit the data, the fitting results highly depend on the fitting condition. for example, the beta value highly depends on the selection of dose interval: https://iopscience.iop.org/article/10.1088/1361-6560/ab369a/meta by Rørvik et al, 2019.

 

R9.

The LQ-model was fitted by the MATLAB_R2019b software using the logarithmic survival data, as shown by the LQ-model equations:

ln S/S0  = −αD − βD2

where D is the physical dose deposited by protons and x-rays, S/S0 is the survival fraction of the radiated cells compared to the reference cells which are not irradiated and α and β are the radiosensitivity parameters. The fit provides the linear and quadratic parameters with their standard errors and the R^2 value. 

The LQ model is the best mathematical representation of cellular radiosensitivity, as proven by the fact that its alpha and beta parameters have been long used in treatment planning since their in vitro determination well reflects the response to radiation of both cancer and normal cells  [Steffen U. et all. DOI 10.1186/s13014-016-0584-z; W. Tinganelli et all. DOI: 10.1038/srep17016; Petković doi10.1080/09553002.2019.1549753; O. Keta doi10.1177/1535370216669611]. The radiobiological interpretation of the beta parameters is related to the cell lines' specific ability to repair sub-lethal radiation-induced damage.

 

Line 213: 3.6x107.Please correct this.

 

R10.

We inserted the suggested corrections (new line 226)

 

Line 219: Please correct this.

 

 

 

R11.

According to the suggestion we modified the sentence as follows (new lines 231-232): “RBE in a clinical setting: microdosimetric-based, Monte Carlo and LEM II model. All of the  proposed approaches were compared with experimental data.”

 

Line 222: why middle? Which position is 'middle'?

 

R12.

The text was improved (new line 235) reporting the depth and particles energy value.

 

Line 228: Please see comment before. Usually OAR is near the distal edge of SOBP, where the LET is much higher. This is also one of the main concerns of using a fixed RBE value.

 

R13.

In order to evaluate the damage induced by proton therapy in a typical tissue belonging to the same anatomical site, the human normal retinal pigment epithelial ARPE-19 cell line was also irradiated.

 

we modified and implemented the text (lines from 240 to 246)

 

Figure 2: Please improve the figure quality Please add legend.  I don't understand if there is only 1 black line or 2?

 

R14.

We inserted the suggested corrections and now Fig 2 is represented by Table 2

 

Figure 3:  please add legend please add the fitting curves of all fittings

 

Figure 4: Same as before

 

R15.

We inserted the suggested corrections and now Fig 3 is represented by Fig 2 and Fig4 represented by Fig 3

 

Line 282: Table 3 is about alpha?

 

R16.

We corrected the text.The sentence is related to Table 4 instead of Table 3.

 

Line 286-287-288: Only one position is tested, which is not enough to support the statement.

 

R17.

The statement has been modified (new line 310) , so that it is now fully supported by the results here presented

 

Line 299-300: only one position in the mid of SOBP is tested. This is not strong enough to support such a conclusion

 

R18.

The statement has been modified (new lines 323-324), so that it is now fully supported by the results here presented

 

Line 308-312: agreement between models doesn't mean those models can predict the ground truth of RBE (i.e. experimental measurement).

 

R19.

We modified the sentence in accordance with the suggestions (new lines 332-335)

Author Response File: Author Response.pdf

Reviewer 3 Report

In this paper, the authors compare experimental results from in vitro cells proton irradiation with Local Effect Model prediction, Geant4 simulations coupled with LEM and micro-dosimetric calculations.

The topic is very interesting and of real importance, being the proton therapy increasingly used. However, the author could stress which is the novelty with respect to the results published in 10.1088/1361-6560/abc368 and 10.1016/j.ejmp.2019.01.018. 

Few minor issues:
-  The second sentence in Results and Discussion is not finished. 
-  Figure 2 should be re-editing with a better resolution

Author Response

We would like to thank the Reviewer for his/her valuable concerns and suggestions.

We implement our work basing on reviewer’s suggestion that are summarized in the following point by point response:

 

Comments and Response:

In this paper, the authors compare experimental results from in vitro cells proton irradiation with Local Effect Model prediction, Geant4 simulations coupled with LEM and micro-dosimetric calculations.

 

The topic is very interesting and of real importance, being the proton therapy increasingly used. However, the author could stress which is the novelty with respect to the results published in 10.1088/1361-6560/abc368 and 10.1016/j.ejmp.2019.01.018. 

 

R1.

We thank Reviewer #3 for his/her constructive inputs. Moreover, we agree with his/her comment and concerns about the fact that cells were exposed solely at one position along the clinical proton SOBP unique position in the beam line (precisely, at which had a depth of 24 mm).The current study pinpoints a clinical and critical aspect regarding the  treatment of neoplasms such as the uveal melanoma by proton therapy. Hence, the correct evaluation of the damage induced by ionizing radiation during RT treatment, also thanks to Monte Carlo-based simulations, represents a valuable and potent tool to foresee the impact of ionizing radiation on cancer tissues. On the other side, it’s equally important to evaluate biological damage induced by an erroneous dose distribution during a proton therapy treatment plan that may have detrimental effects on normal tissue, for such reason the effects on the normal retinal pigment epithelial ARPE-19 cell line were investigated and compared to the ones induced in the uveal melanoma 92.1 cell line. Moreover, as highlighted by recent work by Anfuso et al (2020) and Tosi et al (2021), the uveal melanoma microenvironment is thought to play a decisive role in determining the metatastatic potential of this neoplasm, which very often develops adjacent to the retina and has a high infiltrating capacity. Therefore, the uveal tumour microenvironment is a complex and intricate milieu possibly comprising normal retinal cells.  Irradiating uveal melanoma cancer cells at the mid-SOBP position, as performed in our work, often therefore means irradiating normal retinal cells as well, the radioresponse of which is currently little known.  Obviously, future work will have to address the impact of normal retinal tissue exposure at the distal proton SOBP. Nevertheless, we believe that this is an important starting point to evaluate in vitro the agreement between microdosimetry-based calculations and biological effects in a realistic therapeutic scenario.

 

We modified the text according to the suggestions (new lines 60-72)

 

Few minor issues:

-  The second sentence in Results and Discussion is not finished. 

R2.

The text was modified (new line 233)

-  Figure 2  should be re-editing with a better resolution

R3.

We inserted the suggested corrections and now Fig 2 is represented by Table 2 with a higher resolution

 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

This paper reports the characteristics under proton beam irradiation using two types of cells and two types of detectors, and calculation of radiobiological effects for proton radiotherapy. This manuscript has been improved from previous one. There are some minor comments that need attention in a revised version.

 

1. In the equation (8), does "x" mean a multiplication symbol?

Or please defined the “x”

 

2. Line189: Parameter, RBE_mu is not shown in equation.

Does " RBE_mu " mean RBEµ?

 

3. r(y) is not defined.

 

4. Author used * asterisk for multiplication symbol.

Please change to “×” multiplication sign.

 

5. Is Table2 figure?

 

 

6. In this calculation of RBEµ from TEPC and MicroPlus Bridge microdosimeter, loncol’s function was used. However, this function may be dependent on Cell type as shown in a report “Alessio Parisi et al 2020 Phys. Med. Biol. 65 235010” Please discussed this uncertainty on calculation in this manuscript. And please cite “Alessio Parisi et al 2020 Phys. Med. Biol. 65 235010”

 

 

 

Author Response

We would like to thank the Reviewer number 1 for his/her valuable concerns and suggestions.

We implemented our work basing on reviewer’s suggestion that are summarized in the following point by point response:

This paper reports the characteristics under proton beam irradiation using two types of cells and two types of detectors, and calculation of radiobiological effects for proton radiotherapy. This manuscript has been improved from previous one. There are some minor comments that need attention in a revised version.

Comments and Response:

1. In the equation (8), does "x" mean a multiplication symbol?

Or please defined the “x”

R1.

It’s a typo. We implemented the suggested correction in the equation (8)

2. Line189: Parameter, RBE_mu is not shown in equation.

Does " RBE_mu " mean RBEµ?

R2. It’s a typo. We implemented the suggested correction in the line 205

3. r(y) is not defined.

R3.

We modified the text accordingly. We defined r(y) both in lines 187 and 190.

4. Author used * asterisk for multiplication symbol.

Please change to “×” multiplication sign.

R4.

We implemented the suggested correction in the following lines: 119, 140, 229

5. Is Table2 figure?

R5.

Figure 2 is named Table 2 because to upload the four figures with a higher resolution we created a Table in latex. To maintain the definition of “figure” the resolution would have been too low and that was an issue discussed in the previous round of revision.

6. In this calculation of RBEµ from TEPC and MicroPlus Bridge microdosimeter, loncol’s function was used. However, this function may be dependent on Cell type as shown in a report “Alessio Parisi et al 2020 Phys. Med. Biol. 65 235010” Please discussed this uncertainty on calculation in this manuscript. And please cite “Alessio Parisi et al 2020 Phys. Med. Biol. 65 235010”

R6.

The dependence of the biological weighting function r(y) on cell type is intrinsic in the operative definition of the weighting function (the unfolding procedure starts from RBE data, which depend on the biological end-point. The procedure to determine r(y) is given and discussed in the original paper by Loncol et al., and not in the recent paper by Alessio Parisi. This last paper deals with the calculation of an alternative weighting function, based on deconvolution of biological (V79 cells) and microdosimetric data taken from literature.

The Loncol weighting function has been determined for intestinal tolerance assessment by crypt cells regeneration in mice, as correctly mentioned at lines 187-205 of this work. However, previous works [22, 23] have already shown that the same function allows to reproduce the trend of the RBE10 as  a function of depth for different cell lines, therefore for in vitro RBE. A sentence has been added to the text.  Ideally, the RBE should be normalized at one point, for instance at the SOBP-entrance. Unfortunately, this normalization was not possible here, because cell survival was measured only at the mid-SOBP. We decided therefore to normalize the results at the value RBE=1.1 at the entrance, as described in equation (8) of the present paper. This normalization can results in a systematic shift of the RBE values at the different depths, that was not taken into account.

 

The uncertainty in the scaled RBEm is dominated by the uncertainty in the Loncol’s weighting function, which was however not considered. Instead, statistical uncertainties and the uncertainty related to the lineal energy calibration of the microdosimetric spectra were propagated.

As suggested we have improved the text line 197-200

Reviewer 2 Report

The authors have addressed (point-by-point) all comments and questions raised by the reviewer. The manuscript has been improved and is ready for publication.

Author Response

We would like to thank the Reviewer number 2 for his/her previous suggestions and his/her acceptance of our manuscript

Reviewer 3 Report

The article has been improved following the first revision, and only few minor changes are needed.

Please, add the units of measurement for the dose in Table 4 and 5.

Author Response

We would like to thank the Reviewer for his/her valuable concerns and suggestions.

We implement our work basing on reviewer’s suggestion

1. The article has been improved following the first revision, and only few minor changes are needed.

Please, add the units of measurement for the dose in Table 4 and 5.

R1.

We added the units of measurement as required