Introduction
Esophageal cancer (EC) affects more than 450,000 patients globally per year and is associated with a dismal prognosis [
1]. The main histological subtypes of EC are adenocarcinomas (AC) and squamous cell carcinomas (SCC) with different etiologic factors and regional incidence patterns [
1]. The current standard of care encompasses local treatments for early stage EC, often preceded by neoadjuvant therapies and multimodality treatment approaches for more frequently diagnosed locally advanced EC [
2]. Despite recent advances in EC management, prognosis remains poor with estimated 5-year overall survival rates ranging between 15 to 20% [
3]. Conventional photon radiotherapy (RT) is a key treatment modality for EC patients and is used both in the neoadjuvant and definitive setting, often in combination with systemic treatments [
4‐
7]. Due to mediocre local control rates resulting in dismal patient outcome and potential severe RT-associated morbidity, there is a strong demand for improvements in RT approaches [
8‐
10].
In contrast to conventional photon radiotherapy, particle RT offers several promising physical and biological characteristics that may be of benefit for EC patients. Charged particles exhibit an inverse dose-depth profile that allows a deposition of high radiation doses in a predefined target area (Bragg peak) while sparing surrounding healthy tissues and organs at risk. Additionally, particle RT is characterized by a higher linear energy transfer (LET) and hence an increased relative biological effectiveness compared to conventional photon RT depending on the atomic mass of the used particles and the target cells [
11‐
13]. The higher LET leads to more densely spaced and complex DNA damage, especially irreparable DNA double-strand breaks that helps to overcome hypoxia-associated relative radiation resistance often seen in larger tumors [
12]. Clinical benefits of particle beam RT have been demonstrated for several tumor entities [
14‐
17]. However, the influence and potential benefits of particle RT on EC remain largely unexplored both in vitro and in the clinical setting, and the benefits for EC patients are unknown.
The aim of this study was to systematically evaluate the in-vitro effects of clinically available charged particles on the survival, damage response and DNA repair of EC cell lines of different histologies in comparison with photon RT.
Materials and methods
Cell culture
The human esophageal AC cell lines OE19 and OE33 and the human esophageal SCC lines KYSE270 and KYSE410 were purchased from the European Collection of Authenticated Cell Cultures (ECAAC, Public Health England, Salisbury, UK). OE19, OE33 and KYSE410 cells were cultured in RPMI1640 medium (Biochrom, Berlin, Germany), supplemented with 5% fetal bovine serum (Biochrom) and 1% penicillin-streptomycin (Thermo Fisher Scientific, Dreieich, Germany). KYSE270 cells were grown in 1:1 Ham’s F-12 (Biochrom) and RPMI1640 medium with the addition of 2% fetal bovine serium and 1% penicillin-streptomycin. Cells were maintained in a humidified incubator at 37 °C and 5% CO2.
Radiation treatments
Photon (X) irradiation was performed with a linear accelerator (XRAD 320, Precision X-Ray, North Branford, USA) using single doses ranging between 2 and 10 Gy at 320 kV and a dose rate of 1 Gy/min. Proton (1H), carbon (12C) and oxygen (16O) ion irradiation was performed at the Heidelberg Ion Therapy Center using the raster-scanning technique with a spread-out Bragg peak of 35 ± 5 mm. Energy levels and linear energy transfer for 1H, 12C and 16O irradiation were 64.1–70 MeV/u and 6 keV/μm, 122.4–136,9 MeV/u and 101 keV/ μm, 141.4–160.9 MeV/u and 154 keV/ μm, respectively.
Clonogenic survival assays
Cells were plated in cell culture flasks and left to attach for at least 6 h before irradiation. Cell numbers were adjusted depending on radiation quality and dose (OE19: 400–8000, OE33: 400–8000, KYSE270: 500–9000, KYSE410: 300–4000). After treatment, the cells were maintained for 10 to 14 days to allow cells to form colonies. Colonies were fixed with 25% acetic acid (v/v) in methanol and stained with crystal violet solution. Colonies with more than 50 cells were then counted by light microscopy. For each cell line and condition three independent experiments were performed with three replicate samples. The surviving fraction of cells was calculated based on the following formula: (#colonies/#plated cells)treated/(#colonies/#plated cells)untreated. Replicate samples were averaged and the mean and standard deviation (SD) of all independent experiments was plotted against radiation dose. Survival curves were fitted according to the linear-quadratic model and used to calculate the relative biological effectiveness (RBE) according to the formula: (photon dose)10% survival/(experimental irradiation dose)10% survival.
Flow Cytometry
Cells were plated in triplicates 12 h prior to irradiation with biologically equivalent doses (2 and 8 Gy for X and
1H, 1 and 3 Gy for
12C and
16O) according to the mean RBE value of each irradiation modality. After 2, 8, 24, 48 and 96 h the cells were harvested using trypsin/ethylenediaminetetraacetic acid, fixed with paraformaldehyde and permeabilized with ice-cold 70% ethanol. The samples were washed three times with 0.5% bovine serum albumin (BSA) in PBS and then incubated with antibodies diluted in 3% BSA/PBS for 1 h as follows: The 2 and 8 h samples were incubated with 1:20 diluted Alexa Fluor-488-coupled antibody against γH2AX (BioLegend, San Diego, USA), the 48 and 96 h samples were incubated with 1:20 diluted Alexa Fluor-647-coupled antibody against active caspase-3 (Becton Dickinson, Heidelberg, Germany) and the 24 h samples were incubated with both antibodies at the same dilutions. All samples were then stained with 1 μg/mL DAPI (Sigma Aldrich, Taufkirchen, Germany) in PBS and measured with a LSRII flow cytometer (Beckton-Dickinson), recording 10,000 events per sample. Data analysis was performed with FlowJo 7.6.5 software (LLC, Ashland, USA) as reported before [
18,
19]. Cells were gated (front scatter vs. side scatter plot) to exclude debris and cell doublets/aggregates were filtered out (DAPI-A vs. DAPI-W plot). The DAPI-A histogram of single cells was used to gate subG1 cells vs. cells with normal DNA content and to further classify the latter population into different cell cycle stages using the Dean-Jett-Fox model. G1, S and G2/M cells were gated for cell cycle-specific γH2AX measurement (DAPI-A vs. DAPI-W plot). For each population, the median γH2AX intensity was calculated and the following formula was used to calculate the combined γH2AX levels in the whole cell population (corrected for cell cycle-specific DNA content): I
A = I
G1 * G1 + (I
S * S) / 1.5 + (I
G2|M * G2|M) / 2, where I
A, I
G1, I
S, I
G2|M are median γH2AX intensities of all, G1, S, G2/M cells respectively and G1, S, G2|M are the frequencies of cells in the respective cell cycle phase. The median γH2AX levels of each population was divided by the average of the controls for normalization. For apoptosis measurement both the percentage of subG1-positive and active caspase-3-positive cells were gated in the single cell population and subtracted from the average control levels. All experiments were performed with three replicate samples.
Statistical analyses
Statistical comparison of data was performed by two-sided Student’s t-tests using SPSS Statistics 25 software (IBM, Ehningen, Germany). These tests were paired for comparisons of clonogenic survival data between different radiation modalities or between different cell lines. All described data represent mean values and SD.
Discussion
Here, we demonstrated for the first time the differential effects of photon and particle irradiation on esophageal cancer cell lines of different histologies regarding cellular survival, apoptosis induction and DNA double strand break repair.
Clinically available ion beam radiation with protons or heavier ions like
12C or
16O may make use of the particles’ physical advantages like their inverted dose-depth profile, resulting in the ability to increase treatment doses while sparing surrounding organs-at-risk, or their increased RBE values. Indeed, in our dataset, the RBE values for
12C and
16O radiation were determined to be 2.3 and 2.5, respectively. These values compare well with previous RBE calculations of
12C irradiation obtained from animal models that have suggested values around 2.0 compared to photon irradiation [
18,
19]. No data have been published yet concerning RBE values of particle radiation using heavier ions such as
16O radiation in esophageal cancer.
We found significant differences in cellular sensitivities to photon and particle radiation in the two human esophageal SCC cell lines tested, while the survival data were more consistent for the 2 AC cell lines. Histology-dependent radiation responses have been widely reported for esophageal cancer not only in vitro, but even in patients undergoing clinical radiation therapy [
4,
20,
21]. Esophageal SCC is known to be more radiosensitive compared to AC, resulting in a higher rate of pathological complete response (pCR) after neoadjuvant chemoradiotherapy. Albeit pCR rate is higher for esophageal SCC, both SCC and AC esophageal cancer benefit from preoperative chemoradiotherapy. However, factors beyond tumor histopathology may strongly influence radiation responses of individual esophageal cancers not only to photon but especially particle radiation, and indeed, our data suggest a strong heterogeneity of RBE values that were not histology-specific: RBE values for
12C and
16O radiation ranged between 0.7 and 4.3 and 0.8 and 5.0, respectively, with the SCC cell lines exhibiting both the most radioresistant and radiosensitive phenotypes.
Surprisingly, OE33 cells were observed to be more resistant to photon compared to proton irradiation. Additionally, KYSE410 exhibited similar clonogenic survival curves for
1H,
12C and
16O radiation. Resistance to proton and particle irradiation is linked to effective homologous recombination (HR) DSB repair, and previous studies have shown the important role of HR for esophageal AC and SCC such as OE33 and KYSE410 [
22‐
25]. However, further studies are needed to elucidate the mechanisms for the increased resistance to proton and particle irradiation in these cell lines.
All tested types of particle irradiation resulted in incomplete repair of radiation-induced DNA double strand breaks, correlating with a prolonged G2 phase arrest. Cell cycle arrests are usually signs of ongoing attempts of the cell to repair life-threatening DNA lesions such as DSBs, but are not necessarily predictive for the success of DNA repair. Previous publications have demonstrated correlations of residual γH2AX signals detectable beyond 24 h after irradiation with both increased radiation sensitivity and increased DNA repair capacity [
26‐
28]. In line with these reports, both the radiosensitive KYE270 and the strongly resistant KYSE410 cell lines were found to exhibit strong increases in G2 phase cells with elevated γH2AX levels at 24 h after treatment in our dataset. Additionally, response to particle irradiation was found to correlate with the level of apoptosis induction in all tested esophageal cancer samples, and the cell lines most sensitive to
12C and
16O irradiation exhibited the highest levels of caspase-3 induction at 96 h after treatment. Interestingly, caspase-3 activation was higher after
12C and
16O irradiation versus photons in OE19 and KYSE270 cells, despite of isoeffective dosage, while only marginal differences were observed in OE33 cells and slightly smaller levels were seen for KYSE410 cells irradiated with
12C and
16O versus X and
1H. These variations might indicate that other types of cell death contribute to the overall cell killing effect in a cell line- and radiation modality-dependent manner. Previous reports have outlined a correlation between the induction of apoptosis and the response of esophageal cancer cells and patients to radiation therapy [
29‐
31]. However, it remains unclear if apoptosis levels are linked to the prognostic rates of pathological complete remission after irradiation. Considering the increased biological effectiveness of particle irradiation, especially with heavier
12C or
16O beams and the superior physical dose distribution of these particle treatments, it is conceivable that distinct patient subgroups may derive a clinical benefit from
12C or
16O-based radiation therapy. However, the identification of these patient subgroups does not seem to rely on the histology, and additional molecular markers and signatures are warranted to define the individual benefit of each patient to this novel treatment. While this analysis investigated the use of particle radiation alone on esophageal cancer cells in order to quantify the biological effects, clinical treatment algorithms rely on the combination of radiation with systemic treatments including potentially radiosensitizing anti-cancer drugs such as platinum compounds, 5-fluorouracil or paclitaxel [
4,
6,
32]. Therefore, combination treatments remain to be investigated in order to establish a potential use of particle irradiation in clinically established chemo-radiotherapy protocols.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.