Abstract
Background/Aim: To determine the interaction of gemcitabine in chemoradiotherapy with heavy carbon ions in vitro in a mucoepidermoid carcinoma (MEC) cell line. Materials and Methods: The human lymphatic MEC metastasis cell line NCI-H292 was used. The cells were treated with photons, carbon ions, and gemcitabine. Survival fractions (SF), apoptosis, and cell cycle progression were analyzed. A paired two-sided t-test was used. Significance was defined as p<0.05. Results: Cell proliferation assays showed a significant reduction in SF for combined photon chemoradiation versus photons only. The linear-quadratic fits of combined therapy with carbon ion dose of 0 to 2.5 Gy led to reductions of mean 15% in SF. The LD50 (lethal radiation dose required to reduce cell survival by 50%) for carbon ions only was 0.7 Gy and for carbon ions with gemcitabine 0.6 Gy. The LD50 for photons (with gemcitabine) was 2.8 Gy (2.0 Gy) and for carbon ions (with gemcitabine) 0.7 Gy (0.6 Gy), resulting in a relative biological effectiveness at 10% cell survival (RBE10) of 3.0 (2.7). Carbon ions and photons reduced S phase and increased G2/M phase cell distribution. Isolated treatment with gemcitabine as well as combination with photons led to prolonged S phase transit, whereas combined treatment with carbon ions led to early accumulation in G2/M phase. A significant increase in the sub-G1 population as a hint of relevant number of apoptotic cells was not observed. Conclusion: Gemcitabine showed radiosensitizing effects in combination with photons. The combination of gemcitabine and carbon ions had independent additive effects. Carbon ions only had a RBE10 of 3.0, compared to photons only. The combination of gemcitabine, photon, and carbon ions in patients with MEC seems promising and warrants further investigation.
- Mucoepidermoid carcinoma
- radiotherapy
- in vitro
- particle beam therapy
- ion beam therapy
- carbon ion beam therapy
- chemoradiation
- head and neck
Salivary gland tumors account for approximately 5% of head and neck cancers and mucoepidermoid carcinoma (MEC) for approximately 26% of all salivary gland cancers, mainly located in the parotic glands (1-3). MEC can be differentiated into low-, intermediate- and high-risk tumors (3). High-risk MEC show local aggressive growth and tend towards early systemic spread (4, 5). Treatment for MEC includes surgical resection and adjuvant radiotherapy or in case of inoperability, definitive radiotherapy (6). In palliative treatment, the nucleoside antimetabolite gemcitabine is used. Furthermore, gemcitabine showed radiosensitizing effects in other tumors treated with photons and carbon ions (7). To date, the literature on combined chemoradiotherapy with gemcitabine and carbon ion beams in MEC is limited. Furthermore, studies on combined photon and carbon ion-beam therapy in salivary gland tumors - including MEC - showed improved outcome compared to photon only treatments (8-10).
In the current in vitro study, we analyzed the effects of photon and carbon ion beam therapy in combination with gemcitabine on a metastatic mucoepidermoid carcinoma cell line to implement new therapeutic options for this disease in the future.
Materials and Methods
Cell culture. The human lymphatic metastasis cell line NCI-H292 (European Collection of Authenticated Cell Cultures, Catalogue No.: 91091815) was used. Cells were cultured in RPMI 1640 (Biochrom GmbH, Berlin, Germany) containing 10% FCS (fetal calf serum, Biochrom GmbH) and 1% Penicillin/Streptomycin (Gibco, Thermo Fisher Scientific Inc., Waltham, MA, USA) at 37°C with 6% CO2 and mostly passaged at 75% confluence. For the experiment’s preparation, the cells were seeded in 25 cm2 flasks in particular cell counts, depending on the treatment.
Photon irradiation. The seeded cells were incubated in 25 cm2 flasks for 24 h to allow cells to adhere. The cells were then irradiated using 320 keV photons by X-RAD 320 (Precision X-Ray Inc., Madison, CT, USA) with physical doses of 1.0, 2.0, 3.0, 4.0, 6.0 and 8.0 Gy and then incubated for 5 to 6 days.
Heavy ion irradiation. The carbon ion-beam therapy was performed at Heidelberg Ion-Beam Therapy Centre (HIT) using a horizontal beamline and the raster scanning technique. The cells were irradiated with carbon ion-beams (C12). The single layers of cells were irradiated in the middle of the 8 mm-extended Bragg peak. The mean dose weighted linear energy transfer (LET) value was 103 keV/μm. Single physical carbon ion doses of 0.5, 1.0, 1.5, 2.0 and 2.5 Gy were used.
Chemotherapy. Gemcitabine Hexal 40 mg/ml (Hexal AG, Holzkirchen, Germany) was used and diluted to 1 mg/ml with 0.9 % sodium chloride (B. Braun Melsungen AG, Melsungen, Germany) before the cells were incubated for 2 h. A dose-response relationship and the lethal radiation dose required to reduce cell survival by 50% (LD50) for the cell line were determined using concentrations of 0.05 μg/ml to 1.0 μg/ml gemcitabine. For the combined cell survival experiments, the cells were incubated with 0.5 μg/ml gemcitabine, previously determined as the LD50, in 25 cm2 flasks for 2 h. Subsequently, the cell culture medium was renewed, and the cells were irradiated with photons or carbon ions. After 5 to 6 days of incubation, fixation was performed using 70% ethanol.
Cell proliferation. The cells were harvested from 25 cm2 flasks with 1 ml of EDTA (ethylenediamine-tetraacetate)-trypsin at 37°C, which was afterwards neutralized with 1 ml of medium, resulting in a 1:2 dilution. The cell concentration was then determined using a Neubauer counting chamber.
Clonogenic assay. The clonogenic assay was chosen as a measure of the cell line’s response to the applied chemoradiation. As a gold standard, it captures the radiation and chemotherapeutic toxicity by assessing the loss of colony-forming ability as a defined endpoint. Colony formation was examined 5-6 days after treatment. The flasks were washed using 5 ml PBS and incubated using 5 ml methanol-ice vinegar (4:1) at room temperature before being washed with tab water. Afterwards the cells were stained using 5 ml crystal violet (0.1% in aqua bidest), dried for 2 days, optically counted at 10X magnification and the plating efficiency (PE; number of cells building colonies/seeded total cell count) determined. Then, for each treatment the survival fraction (SF) was calculated by relating the plating efficiency of the sample to the plating efficiency of the control (SF=PE sample/PE control).
Cell cycle analysis. The cell cycle distributions of the cell line were examined using flow cytometry following each therapy. At the time points of 0, 5, 10, 15, 20, and 24 h post irradiation and post incubation with gemcitabine the cells were washed using 2.5 ml PBS, incubated using 0.5 ml EDTA-trypsin, and then fixed using 70% ethanol. The fixed cells were centrifuged, washed with PBS and stained in 100 μl 2 mg/ml RNAse (Serva Electrophoresis GmbH, Heidelberg, Germany) using 10 μl 1 mg/ml propidium iodide (Sigma-Aldrich Co., St. Louis, MO, USA) and 890 μl PBS prior to measurement of DNA-content using a FACScan flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA) at 488 nm wavelength. The cells were analyzed using BD CellQuest Pro 4.0.2 (Becton Dickinson) and the cell cycle phases were determined using ModFit LT 3.0 (Verity Software House, Topsham, ME, USA).
Statistical analysis. Statistical analysis of the clonogenic assays results was performed using Student’s t-test, a paired two-sided t-test. Curves and figures were then calculated and generated using SigmaPlot 12.5 (Systat Software Inc., San Jose, CA, USA). Significance was defined as p<0.05. RBE was defined as the ratio of an absorbed dose of carbon ions to an absorbed reference dose of photons resulting in the same biological effect, for example a specific SF of 10 % (RBE10). Raw data was produced in at least three independent experiments.
Results
Clonogenic survival. The following results were based on the cell proliferation assay at 5 to 6 days after radiotherapy. The photon beam experiments showed a significant reduction of approximately 13% in the SF in combined chemoradiotherapy versus radiotherapy alone (p<0.01, Figure 1). The LD50 of photons only was estimated to approximately 2.8 Gy. A 2 h exposure to 0.5 μg/ml gemcitabine prior to radiotherapy resulted in a steeper curve shape (αphotons+gemcitabine=0.292 versus αphotons=0.186, Table I). The LD50 of combined chemoradiotherapy was estimated to approximately 2.0 Gy.
Linear quadratic fits of normalized cell survival after combined photon beam radiotherapy with gemcitabine versus photon beam radiotherapy alone showed significant reductions in the survival fraction (p<0.01).
Mean regression of cell survival according to the calculated sensitivity values of the linear quadratic model. Loss of the curves shoulder in carbon ion-beam treatments due to β values close to zero.
The normalized experiments on carbon ions combined with gemcitabine did not show significant reductions in the SF compared to carbon ions only (p=0.155, Figure 2). However, the linear-quadratic fit showed a steeper curve shape after combination of gemcitabine and carbon ion beams (αC12+gemcitabine=1.1365 versus αC12=1.0033), whereas the quadratic coefficient β of both treatments tended towards zero (Table I). Considering the linear-quadratic fits, carbon ion-beam experiments showed reductions of mean 15% in the SF. The LD50 for carbon ions only was estimated to be 0.7 Gy, and for carbon ions combined with gemcitabine to be 0.6 Gy.
Linear quadratic fits of normalized cell survival after combined carbon ion-beam radiotherapy with gemcitabine versus carbon ion-beam therapy alone did not show significant reductions in the survival fraction (p=0.155).
The radiation experiments comparing photon beams versus carbon ion-beams revealed a significantly greater reduction in cell survival for C12, characterized by a steeper decline in the cell survival curve, as evident from the sensitivity parameter α (Table I, Figure 3). The LD50 for photon treatments was 2.8 Gy, and for carbon ion only treatments 0.7 Gy, resulting in a RBE10 of 3.0.
Linear quadratic fits of normalized cell survival after carbon ion-beam versus photon beam treatments showing significant reductions in the survival fraction (p<0.001). The RBE10 is 3.0.
The study of carbon of the combinations of ion-beams with gemcitabine versus photon beams with gemcitabine showed a significant decrease in SF after carbon ion-beams (Table I, Figure 4). The combined treatment with carbon ion-beams and gemcitabine had a LD50 of 0.6 Gy, and the combined treatment with photon beams and gemcitabine of 2.0 Gy, resulting in a RBE10 of 2.7.
Linear quadratic fits of normalized cell survival after carbon ion-beams with gemcitabine versus photon beams with 0.5 μg/ml gemcitabine showing significant reductions in the survival fraction (p<0.001). The RBE10 is 2.7.
Cell cycle analysis. Different experiments were performed: 2.0 Gy carbon ion-beams +/− gemcitabine and 6 Gy photon beams +/− gemcitabine. Figure 5 shows the cell cycle analysis at 0, 5, 10, 15, 20, and 24 h after treatment.
Cell cycle distribution after photon- and carbon ion-beams. A) G0/G1 phase distribution of cells after 2.0 Gy carbon ion-beams and 6.0 Gy photon beams. B) S phase cell distribution after 2.0 Gy carbon ion-beams and 6.0 Gy photon beams. C) G2/M phase cell distribution after 2.0 Gy carbon ion-beams and 6.0 Gy photon beams.
Both, carbon ion-beam as well as photon beam treatment reduced the S phase (pC12,S,10h<0.01, pPhotons,S,10h<0.001 respectively) and increased the G2/M phase distribution of cells in comparison to controls significantly (pC12/Photons,G2/M,10h<0.01; Figure 5). The distribution of cells in the G1/G0 phase showed a minor initial decrease with a slow increase after approximately 10 h.
In the combined treatment experiments, 2.0 Gy carbon ion-beams and 6.0 Gy photon beams were combined with 0.5 μg/ml gemcitabine. Figure 6 shows the cell cycle analysis at 0, 5, 10, 15, 20, and 24 h after irradiation with prior exposure to gemcitabine. All independent experiments showed a decrease in G0/G1 phase cell distribution beginning after approximately 5 h and with a minimum after approximately 15 h. Furthermore, after photon beam treatment a delayed S phase transition and a delayed G2/M phase onset were observed (Figure 6). In contrast, no significant increase in S-phase fractions was observed after combined carbon ion irradiation; instead, these fractions decreased by 25% after 10 h. compared to the untreated control (pC12+Gemcitabine,S,10h<0.01). Concurrently, there was an early, albeit due to major standard deviation of the untreated control non-significant (pC12+Gemcitabine,G2/M,10h=0.106), early increase in G2/M phase fractions observed at 5 h, reaching a maximum after 20 h post-irradiation (Figure 6).
Cell cycle distribution after gemcitabine incubation and combination with photon- and carbon ion-beams. A) G0/G1 phase cell distribution following treatment with 2.0 Gy carbon ion-beams with 0.5 μg/ml gemcitabine and 6.0 Gy photon beams with 0.5 μg/ml gemcitabine. B) S phase cell distribution following treatment with 2.0 Gy carbon ion-beams with 0.5 μg/ml gemcitabine and 6.0 Gy photon beams with 0.5 μg/ml gemcitabine. C) G2/M phase cell distribution following treatment with 2.0 Gy carbon ion-beams with 0.5 μg/ml gemcitabine and 6.0 Gy photon beams with 0.5 μg/ml gemcitabine.
Radiation induced apoptosis. All experiments with controls, carbon ion-beams +/− gemcitabine and photon beams +/− gemcitabine were analyzed after 72 h for sub-G1 populations as a marker for apoptosis. A significant increase in sub-G1 populations was not observed, therefore specific tests for apoptosis were not performed (Figure 7).
Flow cytometric analysis of sub-G1 population for all treatment combinations after 72 h (control, photon beams +/− gemcitabine and carbon ion-beams +/− gemcitabine).
Discussion
The aim of this study was to evaluate the effectivity of carbon ion-beam radiotherapy with and without gemcitabine compared to photon beam radiotherapy with and without gemcitabine in vitro in MEC. Furthermore, first experiments on the effects of gemcitabine in combination with carbon ion beam and photon beam radiotherapy on the cell cycle of MEC were performed. These in vitro analyses might improve the therapeutic options in radiation oncology by adding evidence for the combination of gemcitabine with carbon ion-beam or photon beam radiotherapy in MEC.
Up to date, in vitro data on carbon ion-beam radiotherapy in MEC is limited and to our knowledge, no reports on the in vitro combination of carbon ions with gemcitabine have been published.
Investigating the radiosensitizing potential of gemcitabine in combination with photon beam radiotherapy in the MEC cell line NCI-H292, Pauwels et al. reported dose enhancement factors (DEF) of 1.08, 1.31, and 1.60 after doses of 25, 50, and 100 μg/ml difluorodeoxyuridine (dFdU), a metabolite of gemcitabine (11). In our experiments, the threshold dose was 0.2 μg/ml gemcitabine and after analysis of SF and dose-effect curves, the fits showed an LD50 for gemcitabine of 0.5 μg/ml. Furthermore, the results reported by Pauwels et al. correlated very well to our dose-effect relationship values of photon beam radiotherapy alone with an LD50 of approximately 2.8 Gy (11).
Pauwels et al. reported clear radiosensitizing effects of the in vitro combination of photon beam radiotherapy and gemcitabine on the MEC cell line NCI-H292 (12). Those effects were reported to depend on gemcitabine concentration with a minimum of 4 nM and incubation time and decreased with the interval to radiation. LD50 between 2.3 to 3.9 Gy was reported (12). In our study, for 2 h pretreatment with 0.5 μg/ml gemcitabine immediately followed by photon irradiation, the LD50 was 2.0 Gy. Compared to photon beams alone, cell survival was reduced, which can also be observed in the linear sensitivity parameter α. For the linear-quadratic fit, the curves showed a 57% steeper gradient, which was also affected by the quadratic sensitivity parameter β. However, the β effect was relatively smaller than the α effect. Finally, in the combined photon beam and gemcitabine experiments in the cell line NCI-H292, gemcitabine seemed to cause supra-additive effects in the sense of improved radiosensitivity with a mean improvement for photon beam doses 0 to 8 Gy of 13%. Examining other tumor entities, Lawrence et al. concluded in 1997 that gemcitabine was a promising radiation sensitizer and for example El Shafie et al. reported sensitizer enhancement ratios of 1.24-1.66 for the combination of gemcitabine with photon as well as carbon ion-beam radiotherapy on pancreatic cancer cell lines (13, 14).
The experiments on the MEC cell line NCI-H292 using carbon ion beams and their combination with gemcitabine showed independent additive effects. The dose-effect relationship of carbon ion-beams alone showed an LD50 of 0.7 Gy, which is plausible due to the higher LET of carbon ion-beams. To date, there are no publications on in vitro experiments using carbon ion-beams and the MEC cell line NCI-H292 for comparison. The dose-effect relationship of the in vitro combination of carbon ion-beams and gemcitabine in the MEC cell line NCI-H292 was first analyzed in this study. The linear-quadratic fit showed an LD50 of 0.6 Gy after 2 h of exposure to 0.5 μg/ml gemcitabine and carbon ion-beam treatment. However, studies on tumor cell lines other than the MEC cell line NCI-H292 reported comparable results (7, 15). In the study by Schlaich et al. the RBE10 for the WiDr cell line (colon cancer) was 3.1 +/− 0.1 for the combination of carbon ion-beams and gemcitabine (7).
Referring the PEC12+gemcitabine to the untreated control (PEControl), the cell survival SFC12+gemcitabine after carbon ion-beam radiotherapy combined with gemcitabine was significantly reduced (p<0.05). In order to evaluate whether this is due to an interaction between chemotherapy and radiotherapy or to the additive effects of chemotherapy, it is meaningful to relate the experimentally determined PEC12+gemcitabine to the unirradiated control cells incubated with gemcitabine (PEGemcitabine). By doing so, the normalized survival fraction (SFC12+Gemcitabine, normalized) can be calculated, which revealed no significant difference of the carbon ion-beam chemoradiation to the isolated carbon ion irradiated cell survival SFC12 (p=0.155). Hence, no interaction between gemcitabine and carbon ion irradiation could be shown. However, examining the normalized linear-quadratic regression of carbon ions alone versus their combination with gemcitabine, the combination resulted in a mean 15% steeper survival curve. Due to parameter β equaling 0, the curve is mainly shaped by linear parameter α. The relation of the linear parameters α of carbon ions+gemcitabine versus α carbon ions alone was 1.1319, which fits to the above mentioned 15% reduced cell survival.
By considering the experimentally determined, normalized cell survival in conjunction with linear-quadratic regression, independent toxicities, or independent additive effects, of gemcitabine and carbon ion therapy could be demonstrated. Other studies showed cell line specific effects. El Shafie et al. reported additive effects for carbon ion-beam chemoradiation in BxPC-3 and Panc-1 cell lines (13). On the other hand, Schlaich et al. showed independent toxicities in the WiDr cell line (7).
In our experiments, cell survival was significantly decreased after carbon ion-beam versus photon beam radiotherapy alone (p<0.001). The corresponding RBE10 of carbon ions for the MEC cell line NCI-H292 was quantified as 3.0. Other reports on carbon ion-beam versus photon beam radiotherapy showed comparable RBE10 values for the monotherapy (7, 15, 16). In our combined chemoradiation treatments, the RBE10 was 2.7. Other publications on combined chemoradiation with carbon ion-beams showed variable results. For example, El Shafie et al. showed a minor radiosensitizing effect for gemcitabine in combination with carbon ions and photons in the pancreatic cancer cell line AsPC-1 (13). However, in their experiments on the pancreatic cancer cell line Panc-1 only additive effects of the combined carbon ion and gemcitabine treatment were observed (13). The study by El Shafie et al. reported RBE10 values were between 2.6 and 3.1, and comparable to ours (13). Supiot et al. reported radiosensitization in multiple myeloma cell lines for chemoradiation with photons, but not high-LET radiation (17). Shiba et al. demonstrated in their experiment on HeLa cells and the chemotherapeutic agent cisplatin a dependence of radiosensitization on the linear energy transfer (LET) of carbon ion irradiation, showing an increase in radiosensitivity in the low LET range (18). This could be attributed to the ability of high-LET irradiation to overcome radioresistance. Similarly, an increase in the radiosensitizing effect with decreasing LET could be conceivable for the combination therapy with gemcitabine in MECs.
Our cell cycle experiments showed an S phase decrease of approximately 20% after 5 h, and an increase after 15 h post radiation with carbon ions and photons. S phase after photons was shifted approximately 10% downwards but was qualitatively comparable. Furthermore, an increase in G2/M phase cell distribution of approximately 23% after 10 h was observed in both radiation modalities, with restitution after 24 h. Therefore, this indicates a delayed entry and progression through the S phase. Comparable behavior was previously reported on other cell lines that was due to radiation effects and their mechanisms of repair (19, 20). For photon radiation, a prolonged, eventually transient, G1 block was reported and for high-LET radiation G2/M arrests (19, 20). In our experiments, the MEC cell line NCI-H292 showed a qualitatively prolonged S phase transition with accumulation in G2/M phase in both radiation modalities. A (transient) G1 block could not be demonstrated in this work.
A gemcitabine mediated radiosensitization with focus on the early S phase, due to a dose-dependent arrest or prolongation of the S phase transit, is reported for different cell lines (7, 11, 16). This radiosensitizing effect and the prolonged gemcitabine mediated entry and transit of the S phase was confirmed in our experiments for the MEC cell line NCI-H292 concerning combined radiation with photons. Regarding the combined treatment with carbon ions, S-phase transit seems to be less prolonged leading furthermore to an early accumulation in G2/M phase. Regarding apoptosis, a significant increase in the apoptosis rate was not expected for gemcitabine, 2.0 Gy carbon ions, 6.0 Gy photons, or their combination (7, 11). Since our results on the sub-G1 phase were comparable, in-depth experiments on apoptosis were not performed.
Conclusion
Our experiments show radiosensitizing effects for gemcitabine in combination with photon beams in the MEC cell line NCI-H292. The combination of gemcitabine and carbon ion-beams had independent additive effects. Carbon ion-beams alone had a RBE10 of 3.0, compared to photon beams. The combination of gemcitabine, photon, and carbon ion-beam treatments in patients with MEC seems promising and warrants further investigations.
Footnotes
Authors’ Contributions
PEH and JD developed and planned the study. TS, FZ and PEH performed the irradiation and were responsible for management of the cells. TS, FZ, PEH and HH participated in writing the manuscript and revising it. TS, FZ, PEH and HH performed data analysis. PEH, JD and HH reviewed all data and statistical analyses. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors declare that they have no competing interests in relation to this study.
- Received May 22, 2023.
- Revision received June 16, 2023.
- Accepted June 19, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).