This experiment was approved by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University Hospital (IACUC approval No. 12-0179). The National Research Council guidelines for the care and use of laboratory animals were observed. Male 5-week-old BALB/c-nu mice were used in this study. The animals were maintained at the Biomedical Research Institute of Seoul National University Hospital, which was approved by the Korean Food and Drug Administration and complies with the regulations and standards of the IACUC of Seoul National University Hospital. The mice were housed under pathogen-free conditions with controlled humidity (40%-60%) and temperature (20°C-24°C). Animals were housed under a 12-hour light/dark cycle, with lights on from 8 AM to 8 PM The mice were kept in an individual ventilated cage system on sawdust bedding. Standard mouse diet and filtered city tap water from standard Perspex drinking bottles were provided ad libitum.

We used A431 cells that originated from human vulvar epidermoid carcinoma. The cell line was purchased from the American Type Culture Collection (Manassas, VA), and cells were maintained in Dulbecco's modified Eagle's medium (Gibco BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (Gibco BRL). Cells were grown in an incubator with a humidified atmosphere of 95% air/5% CO2 at 37.5°C.

Recombinant human EGF was provided by Daewoong Pharmaceutical Company (Seoul, Korea) in the form of powder. The agent was diluted in phosphate-buffered saline at a concentration of 1 mg/mL and stored at –70°C.

A total of 5×10⁵ A431 cells were injected subcutaneously into the right hind leg of the mice. When the tumors reached a volume of 200 mm3 at about 7-9 days after the inoculation, mice were divided into five groups (n=8 for each group). An additional two mice (n=2) that were not inoculated with tumor cells received intraperitoneal EGF for 20 days in each experiment. The animals were monitored for 6 months without other interventions to evaluate potential adverse effects of EGF on major organs. Fig. 1 illustrates the treatment design of the five experimental groups. The control group (group I) received no treatment, while the others received EGF for 6 days, EGF for 20 days, RT (30 Gy/6 fractions [fx], daily), and RT (30 Gy/6 fx, daily) plus concomitant EGF (for 6 days) (groups II, III, IV, and V, respectively) (Fig. 1). EGF was administered by intraperitoneal injection (5 mg/kg) once a day. The injection dose was determined by considering the half-life of the drug, and the feasibility was examined in the preliminary experiments. RT was delivered using 6-MV photon energy (Clinac 6/100, Varian Medical Systems, Palo Alto, CA). The fraction size was 5 Gy/fx and the total RT dose was 30 Gy. A custom-made acrylic device was employed to immobilize the body and the leg tumors. In the RT plus EGF group (group V), EGF was injected several minutes before irradiation. A time interval existed due to positioning of mice on the treatment couch, opening/closing the door of the treatment room, and the beam-on time with manipulation of the treatment machine. Day 1 was defined as the start date of each treatment.

Tumor size was measured every other day using a Vernier caliper by two independent researchers (Y.J.L. and S.-R.J.) until day 23. To determine a humane endpoint, the entire period of observation was stopped before the maximum diameter of a single tumor exceeded 2 cm. Tumor volume was calculated according to the formula 1/2×length×width² (mm³). Mice were sacrificed on days 0, 12, and 23 to obtain paraffin blocks of tumor tissues and major organs, such as liver, lung, and kidney. The relative tumor volume was defined as the ratio between the final volume and the initial volume. The experiments were independently repeated three times.

Five-micrometer-thick, paraffin-embedded tumor sections were cut and deparaffinized in Dako PT Link (Dako North America Inc., Carpinteria, CA) and then stained with hematoxylin and eosin (H&E). For immunohistochemical staining, the antigen retrieval process was performed at 97°C using target retrieval solution. Slides were rinsed with Envision FLEX Wash Buffer (Dako North America Inc.) and washed with diluted water. Endogenous blocking with 3% H₂O₂ was performed for 5 minutes. The primary antibodies, anti-EGFR (1:50, #4267, Cell Signaling, Danvers, MA) and anti-cleaved caspase-3 (1:50, #9661, Cell Signaling) were diluted with antibody diluents (Invitrogen Life Technologies, Carlsbad, CA). Buffer solution was used again, and secondary antibodies of horseradish peroxidase-labeled polymer anti-rabbit were applied. The samples were developed with Dako REAL 3,3'-diaminobenzidine+chromogen (Dako North America Inc.) and treated with Mayer’s hematoxylin. After multi-step dehydration with 95% and 100% alcohol, xylene was used to remove the alcohol.

H&E-stained slides of tumor and major organ tissues of groups I-V were examined. The morphological findings were then analyzed to assess the systemic impact of exogenous EGF. Upon immunohistochemistry analysis, the expression level of EGFR in A431 cells was evaluated, and cleaved caspase-3 was used to analyze EGF-induced apoptosis in tumor tissues. A single pathologist (J.M.K.) reviewed the immunohistochemistry results without prior knowledge of treatment outcome. The intensity of expression was reported in a semi-quantitative manner.

Data were presented as the mean±standard deviation of the relative tumor volume. The mean data of three independent experiments were included. Since the tumor volumes were characterized as longitudinal data with repeated measurements, the differences among the experimental groups were analyzed by linear mixed model analysis. Groups I-III were analyzed together to evaluate the EGF-induced tumor growth suppression, and groups IV and V were compared to investigate enhancement of the radioresponse. The results were considered to be statistically significant if the p-value was less than 0.05. All analyses were performed using SPSS ver. 20.0 (IBM Co., Armonk, NY).

Figs. 2 and 3 show the relative tumor volumes of all experimental groups. Although mice treated with EGF for 6 days showed decreased tumor growth relative to the control group (Fig. 2), the tumor volume of group II (EGF×6) approached the level of the control group on day 23 (p=0.550). In the RT and RT with concomitant EGF for 6 days groups (groups IV and V, respectively), differences in relative tumor volume were observed after emergence of the first minimal value (Fig. 2). Group IV (RT) showed the minimal value on day 9 (0.92±0.05), while the tumor volume of group V (RT+EGF) decreased continuously until day 13 (0.69±0.11). After the minimal points, relative tumor volumes increased in both group IV (RT) and V (RT+EGF). The slope of the tumor re-growth phase was significantly steeper in group IV than V throughout the follow-up period.

Fig. 3 shows differences in relative tumor volume among groups I (no treatment), II (EGF×6), and III (EGF×20). In contrast to group II (EGF×6), the anti-tumor effect of EGF for 20 days was maintained until the end of the follow-up period. On day 23, the mean relative tumor volumes of group I (no treatment), II (EGF×6), and III (EGF×20) were 17.63, 16.48, and 10.64, respectively. The tumor volume of group III (EGF×20) was significantly lower than that of group I (no treatment) and II (EGF×6; p < 0.001 for both comparisons).

Immunohistochemistry of EGFR displayed strong membrane staining in A431 tumor cells (Fig. 4). The intensity and patterns of the EGFR expression did not differ among groups I-V. The EGFR scoring system used in this study is listed in Table 1 [16].

The immunohistochemistry of the cleaved caspase-3 antibody revealed EGF-induced apoptosis in the tumor tissues. The staining intensity of each group is listed in Table 2. The apoptotic activity of cleaved caspase-3 was defined as the proportion of apoptotic cells in the tumor sections on each slide (day 12). Cytoplasmic uptake of the antibody was stronger in groups II (EGF×6) and III (EGF×20) than in group I (no treatment). The immunostaining intensity was also higher in group V (RT+EGF) than in group IV (RT) (Fig. 5).

No abnormal histological findings were observed on H&E slides of the liver, lung, and kidney of sacrificed mice that had been treated with EGF (Fig. 6). Liver sections showed normal hepatic lobules and portal areas with intact perilobular and intralobular systems. In lung sections, normal bronchiolar and alveolar structures were observed, while no abnormalities were observed in the glomerulus, capillary, or renal ductal structures in kidney sections. Evaluation of the additional mice treated with EGF for 20 days in the absence of tumor inoculation revealed no death events, and no abnormal histologic features were observed in major normal organs during the 6 month follow-up period.