Calculating Variations in Biological Effectiveness for a 62 MeV Proton Beam

Prediction of normal tissue complication probability for rat spinal cord tolerance following ion irradiations

Objective.Currently, treatment planning in cancer hadrontherapy relies on dose-volume criteria and physical quantities constraints. However, incorporating biologically related models of tumor control probability and of normal tissue complication probability (NTCP) would help further minimizing adverse tissue reactions, and would allow achieving a more patient-specific strategy. The aim of this work was therefore the development of a mechanistic approach to predict NTCP for late tissue reactions following ion irradiation.Approach.A dataset on the tolerance of the rat spinal cord was considered, providing NTCP (for paresis of at least grade II) experimental data following irradiation by photons, protons, helium and carbon ions, under different fractionation schemes. The photon data were fit by a mechanistic NTCP model with four parameters, called Critical Element Model; this allowed fixing the two parameters that only depend on the tissue features. Afterwards, the two parameters depending on radiation quality were predicted by applying the BIophysical ANalysis of Cell death and chromosome Aberrations biophysical model, for each ion type and dose-averaged linear energy transfer value.Main results.The predicted NTCP curves for ion irradiation were tested against the ion experimental data, by Chi-Square andp-value calculations. The model passed a significance test at 1% for all the datasets, and 5% for 13 out of 16 datasets, thus showing a good predictive power. The Relative biological effectiveness (RBE) was also calculated and compared with the data for the endpoint of NTCP equal to 50%, and a considerable discrepancy with the commonly calculated RBE for cell survival was shown.Significance.This study highlights the importance of considering the endpoint of interest when computing the RBE, through the application of a NTCP model, and it represents a first step towards the development of an approach to improve treatment plan optimization in therapy. To this aim, the approach needs to be extended to other endpoints and to be applied to patients' data.

First application of the BIANCA biophysical model to carbon-ion patient cases

Objective. The main objective of this work consists of applying, for the first time, the BIANCA (BIophysical ANalysis of Cell death and chromosome Aberrations) biophysical model to the RBE calculation for C-ion cancer patients, and comparing the outcomes with those obtained by the LEM I model, which is applied in clinics. Indeed, the continuous development of heavy-ion cancer therapy requires modelling of biological effects of ion beams on tumours and normal tissues. The relative biological effectiveness (RBE) of heavy ions is higher than that of protons, with a significant variation along the beam path. Therefore, it requires a precise modelling, especially for the pencil-beam scanning technique. Currently, two radiobiological models, LEM I and MKM, are in use for heavy ions in scanned pencil-beam facilities. Approach. Utilizing an interface with the FLUKA Particle Therapy Tool, BIANCA was applied to re-calculate the RBE-weighted dose distribution for carbon-ion treatment of three patients (chordoma, head-and-neck and prostate) previously irradiated at CNAO, where radiobiological optimization was based on LEM I. The predictions obtained by BIANCA were based either on chordoma cell survival (RBE surv ), or on dicentric aberrations in peripheral blood lymphocytes (RBE ab ), which are indicators of late normal tissue damage, including secondary tumours. The simulation outcomes were then compared with those provided by LEM I. Main results. While in the target and in the entrance channel BIANCA predictions were lower than those obtained by LEM I, the two models provided very similar results in the considered OAR. The observed differences between RBE surv and RBE ab (which were also dependent on fractional dose and LET) suggest that in normal tissues the information on cell survival should be integrated by information more closely related to the induction of late damage, such as chromosome aberrations. Significance. This work showed that BIANCA is suitable for treatment plan optimization in ion-beam therapy, especially considering that it can predict both cell survival and chromosome aberrations and has previously shown good agreement with carbon-ion experimental data.

Ganetespib selectively sensitizes cancer cells for proximal and distal spread-out Bragg peak proton irradiation

Objective

Hypersensitivity towards proton versus photon irradiation was demonstrated in homologous recombination repair (HRR)-deficient cell lines. Hence, combined treatment concepts targeting HRR provide a rational for potential pharmaceutical exploitation. The HSP90 inhibitor ganetespib (STA-9090) downregulates a multitude of HRR-associated proteins and sensitizes for certain chemotherapeutics. Thus, the radiosensitizing effect of HSP90-inhibiting ganetespib was investigated for reference photon irradiation and proton irradiation at a proximal and distal position in a spread-out Bragg peak (SOBP).

Methods

A549 and FaDu cells were treated with low-dose (2 nM resp. 1 nM) ganetespib and irradiated with 200 kV photons. Proton irradiation was performed at a proximal and a distal position within a SOBP, with corresponding dose-averaged linear-energy transfer (LET_D) values of 2.1 and 4.5 keV/µm, respectively. Cellular survival data was fitted to the linear-quadratic model to calculate relative biological effectiveness (RBE) and the dose-modifying factor (DMF). Additionally, A549 cells were treated with increasing doses of ganetespib and investigated by flow cytometry, immunoblotting, and immunofluorescence microscopy to investigate cell cycle distribution, Rad51 protein levels, and γH2AX foci, respectively.

Results

Low-dosed ganetespib significantly sensitized both cancer cell lines exclusively for proton irradiation at both investigated LET_D, resulting in increased RBE values of 10–40%. In comparison to photon irradiation, the fraction of cells in S/G2/M phase was elevated in response to proton irradiation with 10 nM ganetespib consistently reducing this population. No changes in cell cycle distribution were detected in unirradiated cells by ganetespib alone. Protein levels of Rad51 are downregulated in irradiated A549 cells by 10 nM and also 2 nM ganetespib within 24 h. Immunofluorescence staining demonstrated similar induction and removal of γH2AX foci, irrespective of irradiation type or ganetespib administration.

Conclusion

Our findings illustrate a proton-specific sensitizing effect of low-dosed ganetespib in both employed cell lines and at both investigated SOBP positions. We provide additional experimental data on cellular response and a rational for future combinatorial approaches with proton radiotherapy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13014-022-02036-z.

Healthy Tissue Damage Following Cancer Ion Therapy: A Radiobiological Database Predicting Lymphocyte Chromosome Aberrations Based on the BIANCA Biophysical Model

Chromosome aberrations are widely considered among the best biomarkers of radiation health risk due to their relationship with late cancer incidence. In particular, aberrations in peripheral blood lymphocytes (PBL) can be regarded as indicators of hematologic toxicity, which is a major limiting factor of radiotherapy total dose. In this framework, a radiobiological database describing the induction of PBL dicentrics as a function of ion type and energy was developed by means of the BIANCA (BIophysical ANalysis of Cell death and chromosome Aberrations) biophysical model, which has been previously applied to predict the effectiveness of therapeutic-like ion beams at killing tumour cells. This database was then read by the FLUKA Monte Carlo transport code, thus allowing us to calculate the Relative Biological Effectiveness (RBE) for dicentric induction along therapeutic C-ion beams. A comparison with previous results showed that, while in the higher-dose regions (e.g., the Spread-Out Bragg Peak, SOBP), the RBE for dicentrics was lower than that for cell survival. In the lower-dose regions (e.g., the fragmentation tail), the opposite trend was observed. This work suggests that, at least for some irradiation scenarios, calculating the biological effectiveness of a hadrontherapy beam solely based on the RBE for cell survival may lead to an underestimation of the risk of (late) damage to healthy tissues. More generally, following this work, BIANCA has gained the capability of providing RBE predictions not only for cell killing, but also for healthy tissue damage.

Investigation of DNA Damage and Cell-Cycle Distribution in Human Peripheral Blood Lymphocytes under Exposure to High Doses of Proton Radiotherapy

This study systematically investigates how a single high-dose therapeutic proton beam versus X-rays influences cell-cycle phase distribution and DNA damage in human peripheral blood lymphocytes (HPBLs). Blood samples from ten volunteers (both male and female) were irradiated with doses of 8.00, 13.64, 15.00, and 20.00 Gy of 250 kV X-rays or 60 MeV protons. The dose-effect relations were calculated and distributed by plotting the frequencies of DNA damage of excess Premature Chromosome Condensation (PCC) fragments and rings in the G2/M phase, obtained via chemical induction with calyculin A. The Papworth's u test was used to evaluate the distribution of DNA damage. The study shows that high doses of protons induce HPBL DNA damage in the G2/M phase differently than X-rays do. The results indicate a different distribution of DNA damage following high doses of irradiation with protons versus photons between donors, types of radiation, and doses. The proliferation index confirms the impact of high doses of mitosis and the influence of radiotherapy type on the different HPBL response. The results illuminate the cellular and molecular mechanisms that underlie differences in the distribution of DNA damage and cell-cycle phases; these findings may yield an improvement in the efficacy of the radiotherapies used.

In Vivo Validation of the BIANCA Biophysical Model: Benchmarking against Rat Spinal Cord RBE Data

(1) Background: Cancer ion therapy is constantly growing thanks to its increased precision and, for heavy ions, its increased biological effectiveness (RBE) with respect to conventional photon therapy. The complex dependence of RBE on many factors demands biophysical modeling. Up to now, only the Local Effect Model (LEM), the Microdosimetric Kinetic Model (MKM), and the "mixed-beam" model are used in clinics. (2) Methods: In this work, the BIANCA biophysical model, after extensive benchmarking in vitro, was applied to develop a database predicting cell survival for different ions, energies, and doses. Following interface with the FLUKA Monte Carlo transport code, for the first time, BIANCA was benchmarked against in vivo data obtained by C-ion or proton irradiation of the rat spinal cord. The latter is a well-established model for CNS (central nervous system) late effects, which, in turn, are the main dose-limiting factors for head-and-neck tumors. Furthermore, these data have been considered to validate the LEM version applied in clinics. (3) Results: Although further benchmarking is desirable, the agreement between simulations and data suggests that BIANCA can predict RBE for C-ion or proton treatment of head-and-neck tumors. In particular, the agreement with proton data may be relevant if the current assumption of a constant proton RBE of 1.1 is revised. (4) Conclusions: This work provides the basis for future benchmarking against patient data, as well as the development of other databases for specific tumor types and/or normal tissues.

Investigating the impact of alpha/beta and LETd on relative biological effectiveness in scanned proton beams: An in vitro study based on human cell lines

PURPOSE

A relative biological effectiveness (RBE) of 1.1 is commonly used in clinical proton therapy, irrespective of tissue type and depth. This in vitro study was conducted to quantify the RBE of scanned protons as a function of the dose-averaged linear energy transfer (LET_d ) and the sensitivity factor (α/ß)_X . Additionally, three phenomenological models (McNamara, Rørvik, and Jones) and one mechanistic model (repair-misrepair-fixation, RMF) were applied to the experimentally derived data.

METHODS

Four human cell lines (FaDu, HaCat, Du145, SKMel) with differential (α/ß)_X ratios were irradiated in a custom-designed irradiation setup with doses between 0 and 6 Gy at proximal, central, and distal positions of a 80 mm spread-out Bragg peak (SOBP) centered at 80 mm (setup A: proton energies 66.5-135.6 MeV) and 155 mm (setup B: proton energies 127.2-185.9 MeV) depth, respectively. LET_d values at the respective cell positions were derived from Monte Carlo simulations performed with the treatment planning system (TPS, RayStation). Dosimetric measurements were conducted to verify dose homogeneity and dose delivery accuracy. RBE values were derived for doses that resulted in 90 % (RBE₉₀ ) and 10 % (RBE₁₀ ) of cell survival, and survival after a 0.5 Gy dose (RBE_0.5Gy ), 2 Gy dose (RBE_2Gy ), and 6 Gy dose (RBE_6Gy ).

RESULTS

LET_d values at sample positions were 1.9, 2.1, 2.5, 2.8, 4.1, and 4.5 keV/µm. For the cell lines with high (α/ß)_X ratios (FaDu, HaCat), the LET_d did not impact on the RBE. For low (α/ß)_X cell lines (Du145, SKMel), LQ-derived survival curves indicated a clear correlation of LET_d and RBE. RBE₉₀ values up to 2.9 and RBE₁₀ values between 1.4 and 1.8 were obtained. Model-derived RBE predictions slightly overestimated the RBE for the high (α/ß)_X cell lines, although all models except the Jones model provided RBE values within the experimental uncertainty. For low (α/ß)_X cell lines, no agreement was found between experiments and model predictions, that is, all models underestimated the measured RBE.

CONCLUSIONS

The sensitivity parameter (α/ß)_X was observed to be a major influencing factor for the RBE of protons and its sensitivity toward LET_d changes. RBE prediction models are applicable for high (α/ß)_X cell lines but do not estimate RBE values with sufficient accuracy in low (α/ß)_X cell lines.

First benchmarking of the BIANCA model for cell survival prediction in a clinical hadron therapy scenario

In the framework of RBE modelling for hadron therapy, the BIANCA biophysical model was extended to O-ions and was used to construct a radiobiological database describing the survival of V79 cells as a function of ion type (1  ⩽  Z  ⩽  8) and energy. This database allowed performing RBE predictions in very good agreement with experimental data. A method was then developed to construct analogous databases for different cell lines, starting from the V79 database as a reference. Following interface to the FLUKA Monte Carlo radiation transport code, BIANCA was then applied for the first time to predict cell survival in a typical patient treatment scenario, consisting of two opposing fields of range-equivalent protons or C-ions. The model predictions were found to be in good agreement with CHO cell survival data obtained at the Heidelberg ion-beam therapy (HIT) centre, as well as predictions performed by the local effect model (version LEM IV). This work shows that BIANCA can be used to predict cell survival and RBE not only for V79 and AG01522 cells, as shown previously, but also, in principle, for any cell line of interest. Furthermore, following interface to a transport code like FLUKA, BIANCA can provide predictions of 3D biological dose distributions for hadron therapy treatments, thus laying the foundations for future applications in clinics.

The Radiobiological Effects of Proton Beam Therapy: Impact on DNA Damage and Repair

Proton beam therapy (PBT) offers significant benefit over conventional (photon) radiotherapy for the treatment of a number of different human cancers, largely due to the physical characteristics. In particular, the low entrance dose and maximum energy deposition in depth at a well-defined region, the Bragg peak, can spare irradiation of proximal healthy tissues and organs at risk when compared to conventional radiotherapy using high-energy photons. However, there are still biological uncertainties reflected in the relative biological effectiveness that varies along the track of the proton beam as a consequence of the increases in linear energy transfer (LET). Furthermore, the spectrum of DNA damage induced by protons, particularly the generation of complex DNA damage (CDD) at high-LET regions of the distal edge of the Bragg peak, and the specific DNA repair pathways dependent on their repair are not entirely understood. This knowledge is essential in understanding the biological impact of protons on tumor cells, and ultimately in devising optimal therapeutic strategies employing PBT for greater clinical impact and patient benefit. Here, we provide an up-to-date review on the radiobiological effects of PBT versus photon radiotherapy in cells, particularly in the context of DNA damage. We also review the DNA repair pathways that are essential in the cellular response to PBT, with a specific focus on the signaling and processing of CDD induced by high-LET protons.

Analysis of Radiation-Induced Chromosomal Aberrations on a Cell-by-Cell Basis after Alpha-Particle Microbeam Irradiation: Experimental Data and Simulations

There is a continued need for further clarification of various aspects of radiation-induced chromosomal aberration, including its correlation with radiation track structure. As part of the EMRP joint research project, Biologically Weighted Quantities in Radiotherapy (BioQuaRT), we performed experimental and theoretical analyses on chromosomal aberrations in Chinese hamster ovary cells (CHO-K1) exposed to α particles with final energies of 5.5 and 17.8 MeV (absorbed doses: ∼2.3 Gy and ∼1.9 Gy, respectively), which were generated by the microbeam at the Physikalisch-Technische Bundesanstalt (PTB) in Braunschweig, Germany. In line with the differences in linear energy transfer (approximately 85 keV/μm for 5.5 MeV and 36 keV/μm for 17.8 MeV α particles), the 5.5 MeV α particles were more effective than the 17.8 MeV α particles, both in terms of the percentage of aberrant cells (57% vs. 33%) and aberration frequency. The yield of total aberrations increased by a factor of ∼2, although the increase in dicentrics plus centric rings was less pronounced than in acentric fragments. The experimental data were compared with Monte Carlo simulations based on the BIophysical ANalysis of Cell death and chromosomal Aberrations model (BIANCA). This comparison allowed interpretation of the results in terms of critical DNA damage [cluster lesions (CLs)]. More specifically, the higher aberration yields observed for the 5.5 MeV α particles were explained by taking into account that, although the nucleus was traversed by fewer particles (nominally, 11 vs. 25), each particle was much more effective (by a factor of ∼3) at inducing CLs. This led to an increased yield of CLs per cell (by a factor of ∼1.4), consistent with the increased yield of total aberrations observed in the experiments.

BIANCA, a biophysical model of cell survival and chromosome damage by protons, C-ions and He-ions at energies and doses used in hadrontherapy

An upgraded version of the BIANCA II biophysical model, which describes more realistically interphase chromosome organization and the link between chromosome aberrations and cell death, was applied to V79 and AG01522 cells exposed to protons, C-ions and He-ions over a wide LET interval (0.6-502 keV µm^-1), as well as proton-irradiated U87 cells. The model assumes that (i) ionizing radiation induces DNA 'cluster lesions' (CLs), where by definition each CL produces two independent chromosome fragments; (ii) fragment (distance-dependent) mis-rejoining, or un-rejoining, produces chromosome aberrations; (iii) some aberrations lead to cell death. The CL yield, which mainly depends on radiation quality but is also modulated by the target cell, is an adjustable parameter. The fragment un-rejoining probability, f, is the second, and last, parameter. The value of f, which is assumed to depend on the cell type but not on radiation quality, was taken from previous studies, and only the CL yield was adjusted in the present work. Good agreement between simulations and experimental data was obtained, suggesting that BIANCA II is suitable for calculating the biological effectiveness of hadrontherapy beams. For both V79 and AG01522 cells, the mean number of CLs per micrometer was found to increase with LET in a linear-quadratic fashion before the over-killing region, where a less rapid increase, with a tendency to saturation, was observed. Although the over-killing region deserves further investigation, the possibility of fitting the CL yields is an important feature for hadrontherapy, because it allows performing predictions also at LET values where experimental data are not available. Finally, an approach was proposed to predict the ion-response of the cell line(s) of interest from the ion-response of a reference cell line and the photon response of both. A pilot study on proton-irradiated AG01522 and U87 cells, taking V79 cells as a reference, showed encouraging results.

Study of the Influence of NanOx Parameters

NanOx is a new biophysical model that aims at predicting the biological effect of ions in the context of hadron therapy. It integrates the fully-stochastic nature of ionizing radiation both at micrometric and nanometric scales and also takes into account the production and diffusion of reactive chemical species. In order to further characterize the new framework, we discuss the meaning and relevance of most of the NanOx parameters by evaluating their influence on the linear-quadratic coefficient α and on the dose deposited to achieve 10% or 1% of cell survival, D10% or D1%, as a function of LET. We perform a theoretical study in which variations in the input parameters are propagated into the model predictions for HSG, V79 and CHO-K1 cells irradiated by monoenergetic protons and carbon ions. We conclude that, in the current version of NanOx, the modeling of a specific cell line relies on five parameters, which have to be adjusted to several experimental measurements: the average cellular nuclear radius, the linear-quadratic coefficients describing photon irradiations and the α values associated with two carbon ions of intermediate and high-LET values. This may have interesting implications toward a clinical application of the new biophysical model.

Proximity effects in chromosome aberration induction: Dependence on radiation quality, cell type and dose

It is widely accepted that, in chromosome-aberration induction, the (mis-)rejoining probability of two chromosome fragments depends on their initial distance, r. However, several aspects of these "proximity effects" need to be clarified, also considering that they can vary with radiation quality, cell type and dose. A previous work performed by the BIANCA (BIophysical ANalysis of Cell death and chromosome Aberrations) biophysical model has suggested that, in human lymphocytes and fibroblasts exposed to low-LET radiation, an exponential function of the form exp(-r/r₀), which is consistent with free-end (confined) diffusion, describes proximity effects better than a Gaussian function. Herein, the investigation was extended to intermediate- and high-LET. Since the r₀ values (0.8 μm for lymphocytes and 0.7 μm for fibroblasts) were taken from the low-LET study, the results were obtained by adjusting only one model parameter, i.e. the yield of "Cluster Lesions" (CLs), where a CL was defined as a critical DNA damage producing two independent chromosome fragments. In lymphocytes, the exponential model allowed reproducing both dose-response curves for different aberrations (dicentrics, centric rings and excess acentrics), and values of F-ratio (dicentrics to centric rings) and G-ratio (interstitial deletions to centric rings). In fibroblasts, a good correspondence was found with the dose-response curves, whereas the G-ratio (and, to a lesser extent, the F-ratio) was underestimated. With increasing LET, F decreased and G increased in both cell types, supporting their role as "fingerprints" of high-LET exposure. A dose-dependence was also found at high LET, where F increased with dose and G decreased, possibly due to inter-track effects. We therefore conclude that, independent of radiation quality, in lymphocytes an exponential function can describe proximity effects at both inter- and intra-chromosomal level; on the contrary, in fibroblasts further studies (experimental and theoretical) are needed to explain the strong bias for intra-arm relative to inter-arm exchanges.

New insights in the relative radiobiological effectiveness of proton irradiation

BACKGROUND

Proton radiotherapy is a form of charged particle therapy that is preferentially applied for the treatment of tumors positioned near to critical structures due to their physical characteristics, showing an inverted depth-dose profile. The sparing of normal tissue has additional advantages in the treatment of pediatric patients, in whom the risk of secondary cancers and late morbidity is significantly higher. Up to date, a fixed relative biological effectiveness (RBE) of 1.1 is commonly implemented in treatment planning systems with protons in order to correct the physical dose. This value of 1.1 comes from averaging the results of numerous in vitro experiments, mostly conducted in the middle of the spread-out Bragg peak, where RBE is relatively constant. However, the use of a constant RBE value disregards the experimental evidence which clearly demonstrates complex RBE dependency on dose, cell- or tissue type, linear energy transfer and biological endpoints. In recent years, several in vitro studies indicate variations in RBE of protons which translate to an uncertainty in the biological effective dose delivery to the patient. Particularly for regions surrounding the Bragg peak, the more localized pattern of energy deposition leads to more complex DNA lesions. These RBE variations of protons bring the validity of using a constant RBE into question.

MAIN BODY

This review analyzes how RBE depends on the dose, different biological endpoints and physical properties. Further, this review gives an overview of the new insights based on findings made during the last years investigating the variation of RBE with depth in the spread out Bragg peak and the underlying differences in radiation response on the molecular and cellular levels between proton and photon irradiation. Research groups such as the Klinische Forschergruppe Schwerionentherapie funded by the German Research Foundation (DFG, KFO 214) have included work on this topic and the present manuscript highlights parts of the preclinical work and summarizes the research activities in this context.

SHORT CONCLUSION

In summary, there is an urgent need for more coordinated in vitro and in vivo experiments that concentrate on a realistic dose range of in clinically relevant tissues like lung or spinal cord.

Proximity effects in chromosome aberration induction by low-LET ionizing radiation

Although chromosome aberrations are known to derive from distance-dependent mis-rejoining of chromosome fragments, evaluating whether a certain model describes such "proximity effects" better than another one is complicated by the fact that different approaches have often been tested under different conditions. Herein, a biophysical model ("BIANCA", i.e. BIophysical ANalysis of Cell death and chromosome Aberrations) was upgraded, implementing explicit chromosome-arm domains and two new models for the dependence of the rejoining probability on the fragment initial distance, r. Such probability was described either by an exponential function like exp(-r/r₀), or by a Gaussian function like exp(-r²/2σ²), where r₀ and σ were adjustable parameters. The second, and last, parameters was the yield of "Cluster Lesions" (CL), where "Cluster Lesion" defines a critical DNA damage producing two independent chromosome fragments. The model was applied to low-LET-irradiated lymphocytes (doses: 1-4Gy) and fibroblasts (1-6.1Gy). Good agreement with experimental yields of dicentrics and centric rings, and thus their ratio ("F-ratio"), was found by both the exponential model (with r₀=0.8μm for lymphocytes and 0.7μm for fibroblasts) and the Gaussian model (with σ=1.1μm for lymphocytes and 1.3μm for fibroblasts). While the former also allowed reproducing dose-responses for excess acentric fragments, the latter substantially underestimated the experimental curves. Both models provided G-ratios (ratio of acentric to centric rings) higher than those expected from randomness, although the values calculated by the Gaussian model were lower than those calculated by the exponential one. For lymphocytes the calculated G-ratios were in good agreement with the experimental ones, whereas for fibroblasts both models substantially underestimated the experimental results, which deserves further investigation. This work suggested that, although both models performed better than a step model (which previously allowed reproducing the F-ratio but underestimated the G-ratio), an exponential function describes proximity effects better than a Gaussian one.