Prospective Observational Study of High-Dose Carbon-Ion Radiotherapy for Pelvic Recurrence of Rectal Cancer (GUNMA 0801)

At the National Institute of Radiological Sciences (NIRS), more than 8,000 patients have been treated for various tumors with carbon-ion (C-ion) radiotherapy in the past 20 years based on a radiobiologically defined clinical-dose system. Through clinical experience, including extensive dose escalation studies, optimum dose-fractionation protocols have been established for respective tumors, which may be considered as the standards in C-ion radiotherapy. Although the therapeutic appropriateness of the clinical-dose system has been widely demonstrated by clinical results, the system incorporates several oversimplifications such as dose-independent relative biological effectiveness (RBE), empirical nuclear fragmentation model, and use of dose-averaged linear energy transfer to represent the spectrum of particles. We took the opportunity to update the clinical-dose system at the time we started clinical treatment with pencil beam scanning, a new beam delivery method, in 2011. The requirements for the updated system were to correct the oversimplifications made in the original system, while harmonizing with the original system to maintain the established dose-fractionation protocols. In the updated system, the radiation quality of the therapeutic C-ion beam was derived with Monte Carlo simulations, and its biological effectiveness was predicted with a theoretical model. We selected the most used C-ion beam with αr = 0.764 Gy(-1) and β = 0.0615 Gy(-2) as reference radiation for RBE. The C-equivalent biological dose distribution is designed to allow the prescribed survival of tumor cells of the human salivary gland (HSG) in entire spread-out Bragg peak (SOBP) region, with consideration to the dose dependence of the RBE. This C-equivalent biological dose distribution is scaled to a clinical dose distribution to harmonize with our clinical experiences with C-ion radiotherapy. Treatment plans were made with the original and the updated clinical-dose systems, and both physical and clinical dose distributions were compared with regard to the prescribed dose level, beam energy, and SOBP width. Both systems provided uniform clinical dose distributions within the targets consistent with the prescriptions. The mean physical doses delivered to targets by the updated system agreed with the doses by the original system within ± 1.5% for all tested conditions. The updated system reflects the physical and biological characteristics of the therapeutic C-ion beam more accurately than the original system, while at the same time allowing the continued use of the dose-fractionation protocols established with the original system at NIRS.

Although carbon ion therapy facilities are expensive, the biological effects of carbon ion beam treatment may be better against cancer (and cancer stem cells) than the effects of a photon beam. To investigate whether a carbon ion beam may have a biological advantage over X-rays by targeting cancer stem-like cells, human colon cancer cells were used in vitro and in vivo. The in vitro relative biological effectiveness (RBE) values of a carbon ion beam relative to X-rays at the D10 values were from 1.63 to 1.74. Cancer stem-like CD133(+), CD44(+)/ESA(+) cells had a greater ability for colony and spheroid formation, as well as in vivo tumorigenicity compared with the CD133(-), CD44(-)/ESA(-) cells. FACS (fluorescence-activated cell sorting) data showed that cancer stem-like cells were more highly enriched after irradiation with X-rays than carbon ion at doses that produced the same level of biological efficacy. A colony assay for cancer stem-like cells showed that RBE values calculated by the D10 levels were from 2.05 to 2.28 for the carbon ion beam relative to X-rays. The in vivo xenotransplant assay showed an RBE of 3.05 to 3.25, calculated from the slope of the dose-response curve for tumor growth suppression. Carbon ion irradiation with 15 Gy induced more severe xenograft tumor cell cavitation and fibrosis without significant enhancement of cells with putative cancer stem cell markers, CD133, ESA, and CD44, compared with 30 Gy X-rays, and marker positive cells were significantly decreased following 30 Gy carbon ion irradiation. Taken together, carbon ion beam therapy may have an advantage over photon beam therapy by improved targeting of putative colon cancer stem-like cells. ©2011 AACR