known differences in radiation response, i.e. radiosensitive types such as seminomas and resistant types as chondrosarcomas. We hypothesized that the number of residual cH2AX foci corresponds to the expected tumour radiation sensitivity. In addition, in order to enhance clinical practicability for future studies the data were used for simulations to test the robustness of the method when omitting dose levels.
Materials and methods
Collection and cultivation of patient-derived tumour specimensThe study has been approved by the Ethics Committee of the Medical Faculty of the University of Tübingen (426/2013BO1). All the patients included in the study were untreated prior to surgical procedure. During collection and cultivation tumour tissues were kept in Dulbecco's MEM culture medium supplemented with 2% HEPES, 1% Na-pyruvate, 1% non-essential amino acids, 1% penicillin streptomycin (all Biochrom AG, Berlin, Germany) and 10% FBS (PAN Biotech GmbH, Aidenbach, Germany). Fresh tumour material was retrieved from surgical specimens and placed in 50 ml Falcon tubes (Becton Dickinson International, Heidelberg, Germany) containing 15 ml culture medium. Tumour specimens were transported to the laboratory and subsequently cut manually with the use of surgical forceps (BD027R, B. BRAUN, Aesculap, Tuttlingen, Germany) and scalpels (FEATHER Safety Razor Co., Ltd., Osaka, Japan Number 23) into approximately 2-3 mm slices before being placed in petri dishes (3.5 cm diameter) coated with a 1.5% agarose layer (A9539, Sigma-Aldrich, Germany) containing 3 ml culture media. During all incubation times the petri-dishes were kept in 95% humidified atmosphere at 37 _C and 5% CO2. For the purpose of the study tumour material from a total of 25 patient tumours with 10 different tumour histologies was collected (Table 1; n = 3 seminomas, n = 1 chondrosarcoma, n = 3 urinary bladder carcinomas (Ca), n = 3 colorectal Ca (2 colon Ca + 1 rectum), n = 3 breast Ca, n = 1 hepatocellular Ca, n = 3 renal cell Ca, n = 3 prostate Ca, n = 4 glioblastoma multiforme (GBM) and n = 2 cervix Ca). In one GBM patient, the specimen was removed from the analysis because in the 0 Gy sample there was no viable tumour part (based on morphological criteria and BrdU positivity) and no estimation of the background foci could be performed. Histology and selection of malignant cells for analysis were confirmed by an experienced pathologist (M.S.).Experimental design for evaluation of cH2AX foci in ex vivo irradiated patient-derived tumour specimensThe experimental design has been previously described [45]. Briefly, after initial cultivation for 24 h, the tumour specimens were irradiated typically with 0, 2, 4, 6, 8 Gy single doses (200 kV, RS225 research system Gulmay Medical LTD, Surrey, England; 15 mA; 0.5 mm Cu; dose rate _1 Gy/min). In two tumours, additional doses were delivered to the seminoma sample (#1) doses of 3 and 5 Gy and GBM (#1) 10
