Enhanced radiobiological effects at the distal end of a clinical proton beam: in vitro study

Yoshitaka MATSUMOTO Taeko MATSUURA Mami WADA Yusuke EGASHIRA Teiji NISHIO Yoshiya FURUSAWA In the clinic, the relative biological effectiveness (RBE) value of 1.1 has usually been used in relation to the whole depth of the spread-out Bragg-peak (SOBP) of proton beams. The aim of this study was to confirm the actual biological effect in the SOBP at the very distal end of clinical proton beams using an in vitro cell system. A human salivary gland tumor cell line, HSG, was irradiated with clinical proton beams (accelerated by 190 MeV/u) and examined at different depths in the distal part and the center of the SOBP. Surviving fractions were analyzed with the colony formation assay. Cell survival curves and the survival parameters were obtained by fitting with the linear-quadratic (LQ) model. The RBE at each depth of the proton SOBP compared with that for X-rays was calculated by the biological equivalent dose, and the biological dose distribution was calculated from the RBE and the absorbed dose at each position. Although the physical dose distribution was flat in the SOBP, the RBE values calculated by the equivalent dose were significantly higher (up to 1.56 times) at the distal end than at the center of the SOBP. Additionally, the range of the isoeffective dose was extended beyond the range of the SOBP (up to 4.1 mm). From a clinical point of view, this may cause unexpected side effects to normal tissues at the distal position of the beam. It is important that the beam design and treatment planning take into consideration the biological dose distribution. - INTRODUCTION Proton beam therapy is considered a new yet well-established modality of treatment for cancer and non-cancer diseases around the world [14]. The number of proton therapy facilities in the world, especially in Japan, has increased, and it has doubled within the last 10 years [5, 6]. More than 60 000 patients have been treated with proton beams, and high control rates for localized tumors have been reported [14, 7]. In recent years, advanced proton therapy [e.g. intensity-modulated proton therapy (IMPT)] has been adapted for irregularly shaped tumors, and the effect is beginning to examined by physical fundamental research [5, 6, 8, 9]. The International Commission on Radiation Units and Measurements (ICRU) recommends defining proton therapy doses as the product of the relative biological effectiveness (RBE) and the physical dose of the proton, with its unit as Gy [11, 12]. Recently, most clinical proton facilities have used a constant RBE value of 1.1, meaning that protons are assumed to be 10% more effective than X-rays or gamma-rays at all positions along the depth dose distribution [1114]. The RBE weighting factor of 1.1 was a consequence of several reviews of the available radiobiological data at those instances [12, 15, 16], with most studies determining the RBE in the center of SOBP. However, there is a general consensus that the RBE of protons depends on the position along the penetration depth [1720]. Recent physical simulation results suggest the RBE is not constant and that it depends on many factors such as beam energy, dose, depth, radiation quality, and track structure [12, 2123]. Additionally, modeling studies suggest that there are significant differences between the biologically weighted dose and the absorbed dose distributions for both tumor and normal tissues (using a theoretical variable RBE value to calculate an RBE-weighted proton treatment plan [2426]). Although many studies have measured the RBE of protons, the experimental conditions were very diverse, with respect to differences in beam energy, position along the depthdose distribution, method of calculating RBE, and cells used. In this study, we have determined the RBE at various depths within the SOBP of clinical proton beams with an incident energy of 190 MeV, and have assessed the biological equivalent dose distribution of proton beams. We have also determined the shift of the distal edge of the biological dose compared with the isoeffective dose. MATERIALS AND METHODS Cell cultures A human salivary gland tumor cell line, HSG (JCRB1070: HSGc-C5), was used in this study. The HSG is a standard reference cell line for the intercomparison of RBE among carbon and proton facilities in Japan, and is also used in other countries, including Germany and Korea [25, 2732]. Cells were cultured in Eagles MEM supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 U/ml penicillin and 100 g/ml streptomycin) and incubated under a humidified atmosphere with 5% CO2 and 95% air at 37C. Subcultured cells were harvested and seeded in a chamber slide flask (Lab-Tech SlideFlask 170920, Nunc) at ~1.5 2.0 105 cells/flask with 3 ml of the medium, and incubated in the incubator for 2 d prior to the experiment. The flasks were fully filled with additional medium on the same day or 1 d before the experiment. Irradiation Horizontal proton beams were accelerated up to 190 MeV by an Azimuthally Varying Field (AVF) cyclotron at the NCCHE (National Cancer Center Hospital East) [31]. In this experiment, we used the nozzle designed for the dual-ring scattering method [24] to obtain a flat dose profile and stable dose intensity over the target area. The proton beam was scattered using two thin scatters on the beam line. These scatters made it possible to obtain a flat dose profile over the target area (2.5% over a 2 5 cm2 field). The beam was then cut off using collimators. The profile to the center position of the physical depthdose distribution of the 5 cm-SOBP (from 125 to 175 mmH2O) was less than 7.2% (Fig. 1A). Fig. 1. (A) Depthdose distribution of the spread-out Bragg-peak (SOBP) of the 190 MeV proton beam used in the present experiment. The depthdose measurement was performed in a water phantom. The closed dots show the irradiation position of each cell sample (150, 159, 165, 168, 171, 174, 177, 180 and 183 mmH2O). (B) The cell sample flask was placed in a specially designed polyethylene block (0.98 g/cm3) containing a space to hold it. The thickness of the polyethylene block in front of the flask was chosen to locate the cells at the adequate depth of the spread-out Bragg-peak (SOBP) beam. HSG cells on the bottom of the chamber slide flasks were set in a specially designed polyethylene block (0.98 g/cm3), and the cell surface was placed at the isocenter of the gantry (Fig. 1B). The depths (at 150, 159, 165, 168, 171, 174, 177, 180 and 183 mmH2O) in the beam were selected using polyethylene blocks of various thicknesses placed immediately upstream of the cells. The measurement of the dose and dose-rate was conducted with PTW Markus Chamber (Type 23343; PTW, Freiburg, Germany) and an electrometer (FLUKE35040; Fluke Biomedical, Cleveland, OH). Subsequently, GafChromic EBT film (International Specialty Products, Wayne, NJ) was used for verification. We also measured the dose per monitor unit at the center of the SOBP, and used the average (...truncated)