Search

Aim To assess radiation response using γH2AX assay in surgical specimens from glioblastoma (GB) patients and their corresponding primary gliosphere culture. To test the hypothesis that gliospheres (stem cell enriched) are more resistant than specimens (bulky cell dominated) but that the interpatient heterogeneity is similar. Material and methods Ten pairs of specimens and corresponding gliospheres derived from patients with IDH-wildtype GB were studied. Specimens and gliospheres were irradiated with graded doses and after 24 h the number of residual γH2AX foci was counted. Results Gliospheres showed a higher Nestin expression than specimens and exhibited two different phenotypes: free floating (n = 7) and attached (n = 3). Slope analysis revealed an interpatient heterogeneity with values between 0.15 and 1.30 residual γH2AX foci/Gy. Free-floating spheres were more resistant than their parental specimens (median slope 0.13 foci/Gy versus 0.53) as well as than the attached spheres (2.14). The slopes of free floating spheres did not correlate with their corresponding specimens while a trend for a positive correlation was found for the attached spheres and the respective specimens. Association with MGMT did not reach statistical significance. Conclusion Consistent with the clinical phenotype and our previous experiments, GB specimens show low radiation sensitivity. Stem-cell enriched free-floating gliospheres were more resistant than specimens supporting the concept of radioresistance in stem cell-like cells. The lack of correlation between specimens and their respective gliosphere cultures needs validation and may have a profound impact on future translational studies using γH2AX as a potential biomarker for personalized radiation therapy.

Introduction Preclinical and clinical data suggest that the chemokine pathway governed by SDF-1 and CXCR4 contributes to a resistant phenotype. This retrospective biomarker study aims to explore the specific prognostic value of SDF-1 and CXCR4 expression in locally advanced head and neck squamous cell carcinomas (HNSCC) treated with primary radiochemotherapy (RT-CT). Material and methods Biopsies from 141 HNSCC tumours of the oral cavity, oropharynx and hypopharynx were evaluated for SDF-1 and CXCR4 expression by immunofluorescence. SDF-1 and CXCR4 expression was correlated with clinico-pathological characteristics and outcome after RT-CT. Results Patients with tumours exhibiting overexpression of intracellular SDF-1 and CXCR4 have a higher risk for loco-regional relapse and a worse overall survival after RT-CT (multivariate analysis, hazard ratio 2.33, CI [1.18–4.62], p = 0.02 and hazard ratio 2.02, CI [1.13–3.59], p = 0.02, respectively). Similar results were observed when only the subgroup of HPV DNA negative patients were analysed (hazard ratio 2.23 and 2.16, p = 0.02 and p = 0.01, respectively). Conclusions Our data support the importance of SDF-1 and CXCR4 expression for loco-regional control and overall survival in HNSCC after primary radiochemotherapy. Prospective multivariate validation and further studies into CXCR4 inhibition to overcome radiation resistance are warranted.

Purpose: To establish a clinically applicable protocol for quantification of residual gamma H2AX foci in ex vivo irradiated tumour samples and to apply this method in a proof-of-concept feasibility study to patient-derived tumour specimens. Material and methods: Evaluation of gamma H2AX foci formation and disappearance in excised FaDu tumour specimens after (a) different incubation times in culture medium, 4 Gy irradiation and fixation after 24 h (cell recovery), (b) 10 h medium incubation, 4 Gy irradiation and fixation after various time points (double strand break repair kinetics), and (c) 10 h medium incubation, irradiation with graded single radiation doses and fixation after 24 h (dose-response). The optimised protocol was applied to patient-derived samples of seminoma, prostate cancer and glioblastoma multiforme. Results: Post excision or biopsy, tumour tissues showed stable radiation-induced gamma H2AX foci values in oxic cells after >6 h of recovery in medium. Kinetics of foci disappearance indicated a plateau of residual foci after >12 h following ex vivo irradiation. Fitting the dose-response of residual gamma H2AX foci yielded slopes comparable with in situ irradiation of FaDu tumours. Significant differences in the slopes of ex vivo irradiated patient-derived tumour samples were found. Conclusion: A novel clinically applicable method to quantify residual gamma H2AX foci in ex vivo irradiated tumour samples was established. The first clinical results suggest that this method allows to distinguish between radiosensitive and radioresistant tumour types. These findings support further translational evaluation of this assay to individualise radiation therapy.

Purpose To apply our previously published residual ex vivo γH2AX foci method to patient-derived tumour specimens covering a spectrum of tumour-types with known differences in radiation response. In addition, the data were used to simulate different experimental scenarios to simplify the method. Materials and methods Evaluation of residual γH2AX foci in well-oxygenated tumour areas of ex vivo irradiated patient-derived tumour specimens with graded single doses was performed. Immediately after surgical resection, the samples were cultivated for 24h in culture medium prior to irradiation and fixed 24h post-irradiation for γH2AX foci evaluation. Specimens from a total of 25 patients (including 7 previously published) with 10 different tumour types were included. Results Linear dose response of residual γH2AX foci was observed in all specimens with highly variable slopes among different tumour types ranging from 0.69 (95% CI: 1.14–0.24) to 3.26 (95% CI: 4.13–2.62) for chondrosarcomas (radioresistant) and classical seminomas (radiosensitive) respectively. Simulations suggest that omitting dose levels might simplify the assay without compromising robustness. Conclusion Here we confirm clinical feasibility of the assay. The slopes of the residual foci number are well in line with the expected differences in radio-responsiveness of different tumour types implying that intrinsic radiation sensitivity contributes to tumour radiation response. Thus, this assay has a promising potential for individualized radiation therapy and prospective validation is warranted.

Purpose Squamous cell carcinomas (SCCs) are characterized by moderate radiosensitivity. We have established the human head & neck SCC cell line SKX, which shows an exceptionally high radiosensitivity. It was the aim of this study to understand the underlying mechanisms. Materials & methods Experiments were performed with SKX and FaDu, the latter taken as a control of moderate radiosensitivity. Cell lines were grown as xenografts as well as cell cultures. For xenografts, radiosensitivity was determined via local tumour control assay, and for cell cultures using colony assay. For cell cultures, apoptosis was determined by Annexin V staining and G1-arrest by BrdU labelling. Double-strand breaks (DSBs) were detected by both constant-field gel electrophoresis (CFGE) and γH2AX-foci technique; DSB rejoining was also assessed by in vitro rejoining assay; chromosomal damage was determined by G01-assay. Results Compared to FaDu, SKX cells are extremely radiosensitive as found for both xenografts (TCD 50 for 10 fractions 46.0Gy [95% C.I.: 39; 54 Gy] vs. 18.9 Gy [95% C.I.: 13; 25Gy]) and cell cultures (D 0.01 ; 7.1 vs. 3.5Gy). Both cell lines showed neither radiation-induced apoptosis nor radiation-induced permanent G1-arrest. For DSBs, there was no difference in the induction but for repair with SKX cells showing a higher level of both, slowly repaired DSBs and residual DSBs. The in vitro DSB repair assay revealed that SKX cells are defective in nonhomologous endjoining (NHEJ), and that more than 40% of DSBs are rejoined by single-strand annealing (SSA). SKX cells also depicted a two-fold higher number of lethal chromosomal aberrations when compared to FaDu cells. Conclusions The extreme radiosensitivity of the SCC SKX seen both in vivo and in vitro can be ascribed to a reduced DNA double-strand break repair, resulting from a defect in NHEJ. This defect might be due to preferred usage of other pathways, such as SSA, which prevents efficient endjoining.