Radioresistant glioblastoma stem cells (GSCs) contribute to tumor recurrence and identification of the molecular targets involved in radioresistance mechanisms is likely to enhance therapeutic efficacy. This study analyzed the DNA damage response following ionizing radiation (IR) in 10 GSC lines derived from patients. DNA damage was quantified by Comet assay and DNA repair effectors were assessed by Low Density Array. The effect of RAD51 inhibitor, RI-1, was evaluated by comet and annexin V assays. While all GSC lines displayed efficient DNA repair machinery following ionizing radiation, our results demonstrated heterogeneous responses within two distinct groups showing different intrinsic radioresistance, up to 4Gy for group 1 and up to 8Gy for group 2. Radioresistant cell group 2 (comprising 5 out of 10 GSCs) showed significantly higher RAD51 expression after IR. In these cells, inhibition of RAD51 prevented DNA repair up to 180 min after IR and induced apoptosis. In addition, RAD51 protein expression in glioblastoma seems to be associated with poor progression-free survival. These results underscore the importance of RAD51 in radioresistance of GSCs. RAD51 inhibition could be a therapeutic strategy helping to treat a significant number of glioblastoma, in combination with radiotherapy.

Glioblastoma (GBM) is the most malignant type of primary brain tumor with a very poor prognosis. The actual standard protocol of treatment for GBM patients consists of radiotherapy and concomitant temozolomide (TMZ). However, the therapeutic efficacy of this treatment is limited due to tumor recurrence and TMZ resistance. Recently isolated, glioma stem-like cells (GSCs) are thought to represent the population of tumorigenic cells responsible for GBM resistance and recurrence following surgery and chemotherapy. In addition, MGMT (O6-methylguanine-methyltransferase) methylation is considered as one of the principal mechanisms contributing to TMZ sensitivity of GBM. In this study we have isolated GSCs from 10 adult GBM patients and investigated the relationship between MGMT methylation status and Temozolomide (TMZ) sensitivity of these lines grown either in stem-like or differentiation promoting conditions. Sensitivity to TMZ was significantly associated with MGMT methylation status in cells committed to differentiation but not in stem-like cells. In addition, patients harboring highly methylated MGMT promoters had a longer overall survival. These results reveal the importance of the differentiation process when considering the predictive value of MGMT status in GSCs for clinical response to TMZ.

Approximately 30% of patients with wild type RAS metastatic colorectal cancer are non-responders to anti-epidermal growth factor receptor monoclonal antibodies (anti-EGFR mAbs), possibly due to undetected tumoral subclones harboring RAS mutations. The aim of this study was to analyze the distribution of RAS mutations in different areas of the primary tumor, metastatic lymph nodes and distant metastasis. A retrospective cohort of 18 patients with a colorectal cancer (CRC) was included in the study. Multiregion analysis was performed in 60 spatially separated tumor areas according to the pathological tumor node metastasis (pTNM) staging and KRAS, NRAS and BRAF mutations were tested using pyrosequencing. In primary tumors, intra-tumoral heterogeneity for RAS mutation was found in 33% of cases. Inter-tumoral heterogeneity for RAS mutation between primary tumors and metastatic lymph nodes or distant metastasis was found in 36% of cases. Moreover, 28% of tumors had multiple RAS mutated subclones in the same tumor. A high proportion of CRCs presented intra- and/or inter-tumoral heterogeneity, which has relevant clinical implications for anti-EGFR mAbs prescription. These results suggest the need for multiple RAS testing in different parts of the same tumor and/or more sensitive techniques.

Background High-throughput sequencing has advanced biomarker identification, enabling more targeted cancer treatment. Various gene panels are used to screen for actionable genomic alterations for clinical practice and trial inclusion. Objectives We aimed to determine the proportion of cancer patients with a targetable molecular tumor alteration and their inclusion in related trials using a routine panel, and to compare these proportions with results from a larger panel combined with molecular tumor board (MTB) discussion. Design We analyzed eligibility and inclusion in targeted therapy trials among all cancer patients treated at Poitiers University Hospital who underwent routine NGS. Methods Clinical trials were identified via the French National Cancer Institute (INCa) and ClinicalTrials.gov databases. Inclusion rates were also assessed for patients analyzed with the broader FoundationOne CDx® (FO) panel. Results Among 1456 patients, 835 targetable mutations were identified: KRAS 43.6%, BRAF 19.0%, and PIK3CA 10.8%. Of 179 patients eligible for a dedicated trial, 19 were enrolled (10.6%), mostly at Poitiers University Hospital (84.2%) and predominantly melanoma patients (11/19). For 78.8% of non-included eligible patients, reasons were undocumented. FO analysis increased actionable alteration detection from 34.0% to 64.0%, with trial inclusion rising from 12.0% to 16.0%. Conclusion Although a substantial fraction of patients harbored actionable alterations, trial inclusion remained low at our center. Improving access requires better physician awareness of available clinical trials to optimize patient enrollment.

3577 Background: Anti-epidermal growth factor receptor monoclonal antibodies (anti-EGFR mAbs) are a standard treatment in metastatic colorectal cancer (mCRC) with wild-type (WT) RAS status. However, approximately 30% of patients with RAS WT mCRC are non-responders to anti-EGFR mAbs, possibly due to undetected tumoral subclones harboring RAS mutations. The aim of this study was to analyze the distribution of RAS mutations in different areas of the primary tumor, metastatic lymph nodes and distant metastasis. Methods: A retrospective cohort of 18 patients with a CRC was included in the study. All tumors were tested for KRAS, NRAS (codon 12, 13, 59, 61, 117 and 146) and BRAF^V600E mutations using transparietal sections. Six were KRAS mutant, one NRAS mutant and one BRAF mutant. Multiregion analysis was performed in 60 spatially separated tumor areas according to the pTNM staging, including 49 in primary tumors (submucosa (pT1), musculosa (pT2) and/or subserosa (pT3)), 10 in metastatic lymph nodes and 1 in metastasis. KRAS, NRAS and BRAF mutations were tested using pyrosequencing. Intra-tumoral heterogeneity is defined by the presence of different tumoral subclones within the same tumor mass. Inter-tumoral heterogeneity consists in the presence of different tumor subclones in various tumor sites (i.e. primary tumor, metastatic lymph nodes or metastases). Results: Among the 60 tumor areas, 35 were KRAS mutated (in 11 tumors), 8 NRAS mutated (in 3 tumors) and one BRAF mutated(in 1 tumor). In primary tumors (n = 18), intra-tumoral heterogeneity for RAS mutation was found in 8 patients (44%). Inter-tumoral heterogeneity for RAS mutation between primary tumors and metastatic lymph nodes or distant metastasis (n = 11) was found in 6 patients (54%). Moreover, 5 tumors had multiple RAS mutated subclones in the same tumor (28%). Indeed, we observed the coexistence of different KRAS mutations (n = 3) as well as KRAS and NRAS mutations (n = 4). Conclusions: A high proportion of CRC presented intra- and/or inter-tumoral heterogeneity, which has relevant clinical implications for anti-EGFR mAbs prescription. These results suggest the need for multiple RAS testing in different part of the same tumor and/or more sensitive techniques.