Comparison of the effects of photon versus carbon ion irradiation when combined with chemotherapy in vitro

The human tumour cell lines A549 (lung), LN-229 (glioblastoma), PANC-1 (pancreas) and WiDr (colon) were obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA). The cells were cultured in DMEM supplemented with 10% foetal calf serum (FCS), respectively in RPMI 1640 supplemented with 10% foetal calf serum (PANC-1). They were maintained in culture at 37°C in humidified air with 6% CO₂.

Photon irradiation was performed as single dose exposure to photon doses of 2, 4, 6 and 8 Gy at room temperature. Photons were delivered by a biological irradiator (Precision X-Ray, North Branford, CT, USA). Photons were delivered at 320 kV and 12.5 mA with an averaged dose rate of 1 Gy per minute.

Carbon ion irradiation was performed at the Heidelberg Ion Therapy Centre (HIT) using a horizontal beamline with rasterscanning technique. To obtain clinically relevant parameters in cell irradiation the dose was delivered as an extended Bragg peak with 103 keV/μm (dose averaged LET). The cell monolayers were irradiated in the middle of the extended Bragg peak adjusted by using a 3 cm acrylic shield. Single carbon ion doses of 0.5, 1, 2 and 3 Gy were applied with an averaged dose rate of 0.5 Gy per minute.

All cell lines were treated with chemotherapy concentrations at medium toxicity level (approximately 50% clonogenic survival) for 4 hours followed by a medium change prior to irradiation. This is an exposure scheme firmly established in our laboratory to limit drug exposure time because active clearance mechanisms as present in the whole organism are absent in the culture flask. Of course a direct simulation of the clinical situation is not possible. WiDr cell line was tested for camptothecin (Enzo Life Sciences GmbH, Lörrach, Germany), cisplatin, gemcitabine and paclitaxel (all obtained from the pharmacy of University Hospital Heidelberg), all other cell lines were tested for combination with camptothecin.

For all substances, dose–response-relationships were generated for single-agent treatment. Following drug concentrations were chosen for combination experiments: camptothecin at 80 nM (A549), 75 nM (LN-229), 25 nM (PANC-1) and 100 nM (WiDr), cisplatin at 2 μM, gemcitabine at 70 nM and paclitaxel at 25 nM.

The effects of single and double treatment on cell survival were evaluated using the clonogenic growth assay representing the radiobiological gold standard. Due to the fact that the final cell death may occur only after additional cell divisions the clonogenic survival is an important marker of cell survival after treatment with ionizing irradiation and antitumour reagents. All cell lines were grown under standard conditions. For analysing the effects of photon or carbon ion irradiation alone or in combination with chemotherapy on clonogenic survival, increasing numbers of cells were plated in 25 cm² flasks (Sarstedt, Nümbrecht, Germany). After an attachment period cells were exposed to specific treatment and kept in the incubator for 10–14 days after which they were stained with crystal violet (Serva Electrophoresis GmbH, Heidelberg, Germany). Colonies were counted by microscopic inspection and only colonies with at least 50 cells were counted. Plating efficiencies and survival fractions were calculated. Each experiment was performed in triplicates and was repeated at 3 different days. Mean values and standard deviations were only calculated from independent experiments. Analysis of statistical significance was performed using the t-test. In combination experiments an approach was invoked utilising the criteria of additivity described by Steel and Peckham [11]. The following definitions were used for the description of drug-radiation interactions: independent toxicity denotes the effect of the combination of two agents where total effect is a product of the effects of single agents. This mechanism needs no interaction between drugs and radiation, or their biological effects. The term additivity and supra-additivity according to Steel and Peckham (see above) is defined for a range of combination toxicities. This range is restricted by the dose–response curve of radiation and the isoeffect curve of radiation on the effect level of chemotherapy, alone. Dose–response curves located within this range represent additive effects, when exceeding this range they describe supra-additive effects (or synergism).

Treatment dependent changes in cell cycle distribution were analysed at 4 different time points: at the beginning of chemotherapy treatment, directly post-irradiation, 8 and 12 hours post-irradiation. After trypsinization the cells were washed and fixed in 70% ethanol. Fixed cells were centrifuged, washed and incubated in RNAse and propidium iodide prior to measurement of DNA-content using a FACScan flow cytometer (Becton Dickinson, Heidelberg, Germany) [12]. Each experiment was repeated at 3 different days. Mean values and standard deviations were only calculated from independent experiments.

Apoptosis was monitored at 12, 24 and 36 hours post-irradiation by Nicoletti-assay [13]. Trypsinized cells were washed and fixed in 70% ethanol. Fixed cells were centrifuged, washed and incubated in a hypotonic fluorochrome solution (PI 50 μg/ml in 0.1% sodium citrate plus 0.1% Triton X-100) in the dark overnight before flow-cytometric analysis. Each experiment was repeated at 3 different days. Mean values and standard deviations were only calculated from independent experiments. Qualitative verification was performed using DAPI staining and microscopic inspection.

The clonogenic survival was calculated for all used cell lines for increasing doses of photon or carbon ion irradiation. All cell lines exhibited the well-known increase of killing efficacy per unit dose of carbon ion exposure leading to a steeper dose–response-relationship for carbon ions compared to photons. The strongest effect was seen for A549 cell line.

From these values we calculated the relative biological efficiencies at 10% survival level for each cell line (Table 1). These RBE₁₀ values range from 2.3 to 3.7 depending on the cell line.

Combination experiments of 4 chemotherapeutic drugs camptothecin, cisplatin, gemcitabine and paclitaxel with both radiation qualities for WiDr cell line were performed (Figure 1). For generation of the survival curves within each independent experiment, PE-values measured in triplicate for each treatment condition and the averaged values were normalized to the respective number of untreated sample. Respective PE-values in the combination experiments were normalized to a drug control yielding surviving fractions displayed in the graphical representation. Slightly enhanced photon sensitivity was found in combination with camptothecin, cisplatin and gemcitabine, whereas all other combinations showed only independent toxicities.

For generation of the survival curves of combination experiments of camptothecin for all cell lines the same procedure was performed (Figure 2). Results showed slightly enhanced photon sensitivity for all cell lines in combination with camptothecin, whereas purely independent toxicities were found with carbon ions leading to lower RBE₁₀-values for combination treatment (Table 1).

All cell lines showed the prevailing dose-dependent G2-arrest after photon and carbon ion irradiation while the maximum for carbon ions was not reached in the analysed time. For all cell lines a cell cycle delay after carbon ion irradiation was observed also in combination therapy. In combination with camptothecin a drug-specific halt in S-phase up to 8 hours post-irradiation with transition into the G2-arrest was seen for all cell lines. Combination with gemcitabine showed a drug-specific halt in S-phase for more than 12 hours post-irradiation with delayed occurrence of G2-arrest for WiDr cell line. In combination with cisplatin no specific influence on cell cycle distribution was visible for WiDr cell line. Paclitaxel showed a stronger increase in G2/M-phase fraction in combination compared to irradiation alone (Figures 3, 4).

All cell lines showed no significant induction of apoptosis neither for irradiation alone nor for combination with camptothecin. For WiDr cell line only paclitaxel showed a significant induction of apoptosis but independent of the used radiation quality (Figure 5).