Search

The management of metastatic cutaneous melanoma is conditioned by the identification of BRAF-activating mutations in tumor DNA. Tumor genotyping is usually performed on DNA extracted from tissue samples. However, these invasive samples are rarely repeated during follow-up, and their analysis requires a sample pre-treatment which may take several weeks. Circulating tumor DNA (ctDNA), released into blood by cancer cells, is a good alternative to tissue sampling. ctDNA is not subject to tumor heterogeneity, and can be analyzed rapidly, making possible the detection of mutations in emergency or in patients whose tumor cannot be sampled. ctDNA can also be analyzed repeatedly during follow-up, for postresection minimal residual disease assessment, for therapeutic response monitoring and for early relapse detection.

Simple Summary Microsatellite instability (MSI) assessment has become a major issue in the management of colorectal cancer, with the recent approval of anti-PD1 immunotherapies in MSI-metastatic colorectal cancer. The reference PCR method (MSI-PCR) can be costly, time and tissue-consuming. However, NGS could facilitate the assessment of MSI status while simultaneously screening for targetable oncogenic mutations (KRAS, NRAS, BRAF) for any colorectal cancer, but the algorithms developed to date use a large number of microsatellites that have not been approved by international guidelines and which are generally incompatible with small NGS panels. We present the MEM algorithm, which mimics the interpretation of MSI-PCR data by a human operator to reliably assess MSI status using only five validated microsatellites (BAT-25, BAT-26, NR-21, NR-24 and NR-27). We demonstrated that the MEM algorithm was in perfect agreement with MSI-PCR results, in terms of both MSI status and individual microsatellite status, in a cohort of 146 patients. Abstract Purpose: MEM is an NGS algorithm that uses Expectation-Maximisation to detect the presence of unstable alleles from the NGS sequences of five microsatellites (BAT-25, BAT-26, NR-21, NR-24 and NR-27). The purpose of this study was to compare the MEM algorithm with a reference PCR method (MSI-PCR) and MisMatch Repair protein immunohistochemistry (MMR-IHC). Methods: FFPE colorectal cancer samples from 146 patients were analysed in parallel by MSI-PCR and NGS using the MEM algorithm. MMR-IHC results were available for 133 samples. Serial dilutions of an MSI positive control were performed to estimate the limit of detection. Results: the MEM algorithm was able to detect unstable alleles of each microsatellite with up to a 5% allelic fraction. Of the 146 samples, 28 (19.2%) were MSI in MSI-PCR. MEM algorithm results were in perfect agreement with those of MSI-PCR, at both MSI status and individual microsatellite level (Cohen’s kappa = 1). A high level of agreement was noted between MSI-PCR/MEM algorithm results and MMR-IHC results (Cohen’s kappa = 0.931). Conclusion: the MEM algorithm can determine the MSI status of colorectal cancer samples on a small NGS panel, using only five microsatellites approved by international guidelines, and can be combined with screening for targetable mutations.

The ability of early (first weeks of treatment) ctDNA kinetics to identify primary resistance to anti-PD1 immunotherapies was evaluated with a validation cohort of 49 patients treated with anti‑PD1 for metastatic BRAF or NRAS-mutated melanoma, alone and pooled with the 53 patients from a previously described derivation cohort. BRAF or NRAS mutations were quantified on plasma DNA by digital PCR at baseline and after two or four weeks of treatment. ctDNA kinetics were interpreted according to pre-established biological response criteria. A biological progression (bP, i.e., a significant increase in ctDNA levels) at week two or week four was associated with a lack of benefit from anti-PD1 (4-month PFS = 0%, 1‑year OS = 13%, n = 12/102). Patients without initial bP had significantly better PFS and OS (4-month PFS = 78%, 1‑year OS = 73%, n = 26/102), as did patients whose ctDNA kinetics were not evaluable, due to low/undetectable baseline ctDNA (4-month PFS = 80%, 1‑year OS = 81%, n = 64/102). ctDNA detection at first-line anti-PD1 initiation was an independent prognostic factor for OS and PFS in multivariate analysis. Overall, early ctDNA quantitative monitoring may allow the detection of primary resistances of metastatic melanoma to anti-PD1 immunotherapies.

Background and objective: Detection of genomic rearrangements like ALK fusions are mandatory in non-small cell lung cancer (NSCLC) as those alterations can be targeted by an increasing number of drugs. Fluorescence in-situ hybridization (FISH) is the gold standard but multiplexed technologies allow the analysis of several targets simultaneously. We have evaluated a novel RT-PCR based assay for the detection of ALK, ROS1 and RET rearrangement in NSCLC patients in this research study. Materials and Methods: FFPE tissue sections from 309 patients with late stage NSCLC were screened for ALK and ROS1 status using immunohistochemistry (IHC) and confirmed with FISH. Total nucleic acids were extracted from tissue sections using the Maxwell RSC RNA FFPE kit (Promega). Fusions were detected by RT-PCR using a prototype ALK/RET/ROS1 Fusion Panel assay (Roche). A subset of discordant cases were further analyzed by RNA sequencing using the QIAseq Targeted RNAscan Lung Cancer Panel (Qiagen). Results: All 309 samples were tested by RT-PCR with a 0% invalid rate. ALK fusions were detected in 39 samples: 29 were ALK FISH+, 5 were ALK FISH not contributive, 3 were ALK FISH- (less than 15% tumor cells presenting a gene fusion), and 2 were ALK IHC- and not tested by FISH. ROS1 fusions were detected in 12 samples: 9 were ROS1 FISH+, 1 was ROS1 FISH-, and 2 were ROS1 IHC- and not tested by FISH. RET fusions were detected in 3 samples. The remaining 255 samples were RT-PCR-: 5 were ALK FISH+ containing fusions not covered by the RT-PCR assay, 1 was ALK FISH+ containing ALK polysomy, 1 was ROS1 FISH+ containing a fusion not covered by the RT-PCR assay, 1 was ROS1 FISH+ containing ROS1 amplification, 1 was ROS1 FISH+ but ROS1 RNAseq-, and 246 samples were IHC- or IHC+/FISH-. The overall concordance rate between RT-PCR and FISH for ALK and ROS1 gene rearrangements was 91.9%. Although the RT-PCR assay did not detect 5 ALK and 1 ROS1 fusions due to design limitations, it did identify the presence of fusions in 9 tumors that were undetected by FISH (8 ALK and 1 ROS1). Conclusion: Using limited amount of biological material, RT-PCR was able to detect a substantial majority of ALK and ROS1 fusions identified by FISH, as well as fusions that were missed by the initial IHC/FISH screening. This study demonstrates RT-PCR as a feasible approach to detecting fusions in NSCLC. Citation Format: Marc G. Denis, Audrey Vallée, Christine Sagan, Guillaume Herbreteau, Sandrine Charpentier, Jaya Rajamani, Michael Lee, Ellen Ordinario. Detection of ALK, ROS1 and RET fusion transcripts in FFPE samples of non-small cell lung cancer patients using a novel RT-PCR based assay [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1271.

Circulating tumour DNA (ctDNA) can be used to identify gene alterations. The purpose of this study was to determine whether the detection of ctDNA, based on the identification of BRAF and NRAS mutations before systemic treatment initiation, was associated with the prognosis of metastatic melanoma. In total, 68 BRAF or NRAS-mutated stage IV or unresectable stage III metastatic cutaneous melanoma patients were included and tested for the presence of BRAF and NRAS mutations in circulating DNA before treatment initiation, using the Cobas BRAF/NRAS Mutation Test (Roche). The expected mutation was detected in the plasma of 34/68 patients (50% sensitivity). ctDNA detection was associated with AJCC stage, along with the number and nature of metastases. ctDNA was less frequently detected in NRAS-mutated than in BRAF-mutated melanoma (36% and 66%, respectively). At initiation of first-line treatment, ctDNA detection was associated with poor prognosis in Progression Free Survival (PFS) and Overall Survival (OS) in univariate analysis (log-rank: p = 0.002 and p < 0.0001, respectively). In multivariate analysis, ctDNA detection was an independent factor of poor prognosis in OS, after adjustment for AJCC stage, number and nature of metastases and gender (HR = 4.384; 95% CI: (1.308; 14.699); p = 0.017).

Background and objective: Detection of genomic rearrangements like ALK fusions are of great interest in non-small cell lung cancer (NSCLC) as those alterations can be targeted by an increasing number of drugs. To overcome tissue limitations, detection of these alterations from liquid biopsies is an unmet need, despite the development of novel NGS-based tests. To allow the detection of ALK rearrangements from circulating-free RNA (cfRNA) from NSCLC patients, we have evaluated a novel RT-PCR based assay and compared the results to tissue-based testing using immunohistochemistry (IHC) or fluorescence in-situ hybridization (FISH). Materials and Methods: Sixty-five patients with late stage NSCLC were included in the study. ALK status was assessed on tissue section using immunohistochemistry and/or FISH. cfRNA was extracted from 2 ml of plasma from EDTA or Streck BCT DNA tubes using a prototype cfRNA Sample Preparation method (Roche Molecular Systems, Pleasanton, CA). For the detection of ALK-rearrangements from plasma, a prototype ALK/RET/ROS1 Fusion Panel assay (Roche) was used. Results: The analytical sensitivity of the assay was evaluated by testing plasma from healthy individuals spiked with RNA extracted from FFPE tissue section of an ALK positive tumor. Positive results were obtained with samples spiked with as little as 3 ng RNA. Of the forty-five ALK-fusion positive patients, 11 patients were already under anti-ALK therapy and in none of the patients an ALK fusion could be detected. However, of five samples tested at progression, two (40%) were tested positive. For the 29 ALK-fusion positive samples which were included at baseline prior to therapy, 11 samples have been tested positive for ALK fusions (38%). All 20 negative controls were negative using the PCR based assay. Consequently, the key test parameters for samples tested at baseline were: sensitivity = 37.93% [95% CI 20.69% - 57.74%]; specificity = 100.00% [95% 83.16% - 100.00%]; positive predictive value = 100%; negative predictive value weighted for a prevalence of ALK fusions of 4% in metastatic NSCLC patients = 97.48% [95% CI 96.68 - 98.09]. Conclusion: The prototype cobas ALK/RET/ROS1 Fusion Panel assay was able to detect ALK fusion transcripts in the plasma of NSCLC patients at baseline as well as at disease progression. Limited sensitivity could be explained by biological factors influencing nucleic acid shedding by tumours, as well as the presence of fusions not covered by the assay. However, the assay demonstrated high specificity. These data demonstrate that this assay could potentially be used to select patients for an anti-ALK therapy when tissue samples are not available. Citation Format: Simon Heeke, Marius Ilié, Maryline Allegra, Audrey Vallée, Carole Salacroup, Virginie Tanga, Véronique Hofman, Jaya Rajamani, Michael Lee, Ellen Ordinario, Marc G. Denis, Paul Hofman. Detection of ALK fusion transcripts in plasma of non-small cell lung cancer patients using a novel RT-PCR based assay [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5299.

Major advances in the treatment of non-small cell lung cancer (NSCLC) patients have been obtained during the last decade. Molecular testing of tumor samples is therefore mandatory in routine clinical practice. Tumor DNA is also present as cell-free molecules in blood, which is therefore a very useful and convenient source of tumor DNA. In this review, we discuss pre-analytical and analytical aspects of circulating tumor DNA (ctDNA) analysis. We also describe the use of ctDNA analysis in routine clinical practice, and discuss the potential use of ctDNA monitoring both to identify minimal residual disease and as a potential tool to early identify patients’ response to treatment.

e21695 Background: Detection of genomic rearrangements like ALK and ROS1 fusions are mandatory in non-small cell lung cancer (NSCLC) as those alterations can be targeted by an increasing number of drugs. Fluorescence in-situ hybridization (FISH) has been so far the gold standard but multiplexed technologies would allow analysis of many potential targets simultaneously. We have evaluated a novel RT-PCR based assay and a targeted RNA sequencing solution for the detection of ALK and ROS1 rearrangement in NSCLC patients. Methods: 41 patients with late stage NSCLC were included in the study. ALK and ROS1 status were screened on FFPE tissue sections using immunohistochemistry and confirmed with FISH. Total nucleic acids were extracted from tissue sections using the Maxwell RSC RNA FFPE kit (Promega). Detection of rearrangements was performed by RT-PCR using the prototype ALK/RET/ROS1 Fusion Panel assay (Roche), which detects 7 ALK fusion variants, 13 ROS1 variants and 6 RET variants. For RNA sequencing, the libraries were prepared with the QIAseq Targeted RNAscan Lung Cancer Panel (Qiagen). Sequencing was performed on a MiSeq instrument (Illumina). Results: We tested 16 ALK FISH+ tumors, 4 ROS1 FISH+ tumors, and 21 samples that were either IHC- or IHC+/FISH-. The RT-PCR assay detected 15 of the 16 ALK FISH+ tumors, and 3 of the 4 ROS1 FISH+ tumors. In the 21 IHC- or IHC+/FISH- samples, 1 additional ROS1 fusion and 1 RET fusion were detected. The remaining 19 assays samples were negative. 2 fusions were missed by RT-PCR because the assay design does not cover these rearrangements (KLC1-ALK and CTNND1-ROS1). With targeted RNA sequencing, we were able to detect 14 of the 16 ALK FISH+ tumors and all 4 of the ROS1 FISH+ tumors. We also identified 1 additional ROS1 fusion and 1 RET fusion in the 21 IHC- or IHC+/FISH- samples. 17 patients were negative and 2 tests were not contributive. Overall, using FISH as a reference, the sensitivity of both assays was 90% (18/20). Conclusions: Using limited amount of biological material, both RT-PCR and targeted RNA sequencing were able in one assay to detect the overwhelming majority of ALK and ROS1 fusions identified with FISH. These approaches also allowed us to detect gene rearrangements that were missed by the initial IHC/FISH screening. The RT-PCR assay requires less RNA (as little as 3 ng) and has a much faster turn-around time. The RNA sequencing approach allowed us to identify the fusion partners, but it requires much more RNA and is rather difficult to implement in routine practice.

The management of metastatic melanoma has evolved since the onset of treatments with BRAF inhibitors. In order to predict which patients are likely to respond to these treatments, the therapeutic strategy is now conditioned by the search for the activating mutations of the BRAF gene. Tumor genotyping is routinely performed from DNA extracted from tissue or cellular specimens from the primary tumor, metastases, or neoplastic effusions. Due to their invasiveness, these specimens are rarely repeated during the management. In addition, the analysis of the tumor material requires a pretreatment of the sample (formalin fixation, paraffin inclusion, preparation of tissue sections) and may take up to several weeks, making emergency treatment with BRAF inhibitors impossible. Circulating tumor DNA (ctDNA), released by cancer cells in the blood stream, appears as an alternative to tissue sampling. The pre-analytical conditions are now well defined, and several technological approaches can be used to demonstrate the desired molecular alterations. ctDNA is less affected by tumor heterogeneity, can be collected in a minimally invasive manner and analyzed rapidly. Furthermore, ctDNA can be repeatedly analyzed during follow-up, which makes it possible to envisage its use as a specific tumor marker, in order to monitor the response to the treatment and to detect treatment failure.

To the Editor: The improvement of sensitive techniques has allowed tumor-specific aberrations in circulating tumor DNA (ctDNA)1 to be detected, including mutations in oncogenes and tumor suppressor genes, microsatellite instability, and DNA methylation (1). Therefore, ctDNA is a potential alternative source of tumor DNA that could be used for noninvasive identification of biomarkers at various time points during the course of the disease. In this context, digital PCR has been used to detect BRAF (B-Raf proto-oncogene, serine/threonine kinase) mutations in ctDNA from melanoma patients treated with a BRAF inhibitor (2), showing that BRAF mutation detection in ctDNA is useful as a follow-up to treatment response in patients with advanced melanoma (2). Sensitive techniques such as allele-specific amplification, digital PCR, and next-generation sequencing are used in most studies, but preanalytical procedures are also important. It is now clear that plasma should be preferred (3). Blood samples should be processed in the hours immediately after collection to prevent release of genomic DNA from white blood cells, which …

Background Activating mutations of the EGFR gene in lung carcinoma are associated with response to tyrosine kinase inhibitors. Therefore, a rapid, sensitive assay for mutation detection using routine pathological specimens is mandatory in clinical practice. Methods We have compared our in-house procedure to the Therascreen® EGFR RGQ PCR kit (Qiagen). This assay, based on allele-specific amplification, is approved in the United States, in Europe, Japan and China. Results We first selected a series of 209 that were representative of our routine practice during the last 2 years. Using our assays, EGFR mutations were detected in 36 (17.4%) of these patients (18 p.L858R mutations and 18 exon 19 deletions). All these alterations were also detected using the Therascreen® kit. In addition, this kit allowed us to detect 7 additional alterations: one exon 19 alteration (c.2239_2240TT>CC, p.L747P), 3 p.G719X mutations and 3 p.S768L mutations. In the second part of our study, we selected 81 samples that were identified as deleted for exon 19 using our assay. Eighteen different deletions were described following sequencing. All these samples were tested positive with the Therascreen® kit. Conclusion The Therascreen® EGFR RGQ kit was found to be very powerful (sensitivity 100%; specificity 100%) for the detection of the most frequent EGFR alterations that are clearly associated with response to tyrosine kinase inhibitors.

BACKGROUND Progress in the liquid biopsy field, combined with the development of droplet digital PCR (ddPCR), has enabled noninvasive monitoring of mutations with high detection accuracy. However, current assays detect a restricted number of mutations per reaction. ddPCR is a recognized method for detecting alterations previously characterized in tumor tissues, but its use as a discovery tool when the mutation is unknown a priori remains limited. METHODS We established 2 ddPCR assays detecting all genomic alterations within KRAS exon 2 and EGFR exon 19 mutation hotspots, which are of clinical importance in colorectal and lung cancer, with use of a unique pair of TaqMan® oligoprobes. The KRAS assay scanned for the 7 most common mutations in codons 12/13 but also all other mutations found in that region. The EGFR assay screened for all in-frame deletions of exon 19, which are frequent EGFR-activating events. RESULTS The KRAS and EGFR assays were highly specific and both reached a limit of detection of CONCLUSIONS This method presents the advantage of detecting a higher number of mutations with single-reaction ddPCRs while consuming a minimum of patient sample. This is particularly useful in the context of liquid biopsy because the amount of circulating tumor DNA is often low. This method should be useful as a discovery tool when the tumor tissue is unavailable or to monitor disease during therapy.

International audience; Targeted inactivations of RNA-binding proteins (RNA-BPs) can lead to huge phenotypical defects. These defects are due to the deregulation of certain mRNAs. However, we generally do not know, among the hundreds of mRNAs that are normally controlled by one RNA-BP, which are responsible for the observed phenotypes. Here, we designed an antisense oligonucleotide ("target protector") that masks the binding site of the RNA-BP CUG-binding protein 1 (CUGBP1) on the mRNA Suppressor of Hairless [Su(H)] that encodes a key player of Notch signaling. We showed that injecting this oligonucleotide into Xenopus embryos specifically inhibited the binding of CUGBP1 to the mRNA. This caused the derepression of Su(H) mRNA, the overexpression of Su(H) protein, and a phenotypic defect, loss of somitic segmentation, similar to that caused by a knockdown of CUGBP1. To demonstrate a causal relationship between Su(H) derepression and the segmentation defects, a rescue experiment was designed. Embryonic development was restored when the translation of Su(H) mRNA was re-repressed and the level of Su(H) protein was reduced to a normal level. This "target protector and rescue assay" demonstrates that the phenotypic defects associated with CUGBP1 inactivation in Xenopus are essentially due to the deregulation of Su(H) mRNA. Similar approaches may be largely used to uncover the links between the phenotype caused by the inactivation of an RNA-BP and the identity of the RNAs associated with that protein.

CD9, a member of the tetraspanin family, has been implicated in hematopoietic and leukemic stem cell homing. We investigated the role of CD9 in the dissemination of B acute lymphoblastic leukemia (B-ALL) cells, by stably downregulating CD9 in REH and NALM6 cells. CD9 expression was associated with higher levels of REH cell adhesion to fibronectin and C-X-C motif chemokine receptor 4 (CXCR4)-mediated migration. Death occurred later in NOD/SCID mice receiving REH cells depleted of CD9 for transplantation than in mice receiving control cells. After C-X-C motif chemokine ligand 12 (CXCL12) stimulation, CD9 promoted the formation of long cytoplasmic actin-rich protrusions. We demonstrated that CD9 enhanced RAC1 activation, in both REH cells and blasts from patients. Conversely, the overexpression of a competing CD9 C-terminal tail peptide in REH cytoplasm decreased RAC1 activation and cytoplasmic extension formation in response to CXCL12. Finally, the inhibition of RAC1 activation decreased migration in vitro, and the depletion of RAC1 protein from transplanted REH cells increased mouse survival. Furthermore, a testis-conditioned medium induced the migration of REH and NALM6 cells, and this migration was impeded by an anti-CD9 antibody. The level of CD9 expression also influenced the homing of these cells in mouse testes. These findings demonstrate, for the first time, that CD9 plays a key role in the CXCR4-mediated migration and engraftment of B-ALL cells in the bone marrow or testis, through RAC1 activation.

Lung cancer patients carrying sensitive epidermal growth factor receptor (EGFR) mutations show dramatic responses to tyrosine kinase inhibitors (TKIs). However, the majority of patients whose disease responds to drugs eventually develop resistance to these EGFR-TKIs. The T790M gatekeeper mutation in the EGFR tyrosine kinase domain accounts for half of resistance to these drugs. In some patients, this mutation is also detected as a primary event before drug exposure, at a frequency that is highly dependent on the technique used. This review will focus on the methods that have been used to detect the T790M mutation, and its potential clinical applications both in TKI naive patients and in patients with an acquired resistance.

The relationship between BRAF mutations and the patient clinical profile is still under question. The objective of the present study was to correlate the BRAF mutation status in primary and metastatic melanomas with the clinicopathological profile, disease-free (DFS) and overall survival (OS). A total of 367 melanoma samples from 278 patients were screened for their BRAF status using a combination of allele-specific amplification and DNA sequencing. Two or three tissue samples from the same patient were available for 74 patients. The clinicopathological characteristics were tested for their association with the BRAF mutation using the Fisher's or Pearson's χ2 test. Log-rank tests and Cox models were used for survival analyses. BRAF mutation was found in 152 samples (41.4%). Ten of the 74 patients with several tissue samples (13.5%) had discordant BRAF mutation results. BRAF-mutated patients were significantly younger at the time of primary melanoma and first diagnosis of metastasis than BRAF wild-type patients but with no difference in DFS and OS. According to our results, a primary melanoma with BRAF mutation is not associated with a more aggressive illness.

Activating mutations of the epidermal growth factor receptor (EGFR) in lung tumors are associated with a dramatic response to tyrosine kinase inhibitors. Therefore, routine analysis of pathological specimens is mandatory in clinical practice. We have prospectively tested tumors from Caucasian lung tumor patients between January 2010 and June 2012. DNA was extracted from formalin-fixed paraffin-embedded tissues following macrodissection. The p.L858R substitution was assessed by allele-specific PCR and exon 19 deletions by PCR and DNA fragment analysis. Using a robust process from patient sampling to screening methods, we analyzed samples from 1,403 patients. The EGFR status could be successfully determined for 1,322 patients. EGFR mutations were detected in 179 (13.5%) patients, with female and adenocarcinoma histology predominance. Mutated patients were significantly older than non-mutated patients. Similar mutation rates were obtained with primary tumors and metastases, and with surgical resection, bronchial biopsies, CT-guided needle biopsies and transbronchial needle aspiration. The sensitivity of our assays allowed us to detect EGFR mutations in samples poor (

e21026Background: Circulating tumor DNA is an alternative source of tumor DNA that could be used for noninvasive identification of BRAF mutations at various time-points during the course of the dis...

e20545Background: EGFR alterations can be detected in the plasma of lung cancer patients mutated in their tumor. But there has been no prospective report evaluating the efficacy of tyrosine kinase ...

Introduction and hypothesis : B-acute lymphoblastic leukemia (B-ALL) expressing the fusion transcript ETV6/RUNX1 is one of the most frequent subtype of childhood leukemia. This disease is associated with a good survival rate but 5 to 10% of patients relapse more than five years after treatment most notably in extra-hematopoietic sanctuary sites like testis. CD9, a protein member of the tetraspanin family, has been shown to be under-expressed in ETV6/RUNX1 B-ALL. This protein is involved in hematopoietic stem cell engraftment. Its expression has also been correlated with the risk of metastasis and is associated with a poor clinical outcome in various types of cancer. Our hypothesis is that CD9, through its functional properties on migration and homing, could be a key actor of ETV6/RUNX1 leukemia relapse. The purpose of our study was to investigate the effect of CD9 expression on motility and engrafment of pre-B lymphocytes. Methods and results : We used the human pre-B lymphoblastic REH cell line which express CD9. By lentiviral transduction of small hairpin RNAs targeting CD9 mRNA, we generated REH cells depleted in CD9 protein. We performed engraftment assays in vivo using Nod Scid Gamma immunodeficient mice. Mice injected with REH cells depleted in CD9 had a longer survival rate than mice injected with REH control cells. In vitro analysis of adhesion on fibronectin demonstrated that cellular adhesion is dependent of membrane expression of CD9. As well, the more CD9 is expressed, the higher migration rate in response to CXCL12 chemokine is. F‐actin labelling after CXCL12 stimulation showed also an increased F‐actin polymerization in CD9‐positive cells and the formation of actin extensions where CD9 and actin are colocalized. No modifications in CXCR4 (receptor of CXCL12) internalization has been found in CD9 depleted cell lines, which suggests that CD9 does not affect the receptor itself but the signalling pathway downstream the receptor. We investigated the activation of RAC1 protein, a Rho GTPase known to induce the formation of actin-rich lamellipodia in diverse cellular functions including migration. We demonstrated that cells depleted in CD9 have lost their ability to activate RAC1 in response to CXCL12. The inhibition and the depletion of RAC1 respectively have the same effect as CD9 depletion on the migration in response to CXCL12 in vitro and on the survival in vivo. Finally, we tested the ability of REH cells and REH depleted in CD9 cells to migrate to testis in vivo after xenograft in mouse model. We showed that only CD9 positive-REH cells were found in testis tissue after xenograft. Moreover we tested in vitro, the ability of testis tissue to induce the migration of REH cells in a Boyden chamber model. We showed that testis tissue induced the migration of REH cells and this migration was strongly decreased using either an inhibitor of CXCR4, an inhibitor of RAC1 or an antibody targeting specifically CD9. Conclusion : Our data show that CD9 is a key actor in pre‐B lymphoblasts adhesion, CXCL12 dependant migration and actin remodelling. We show for the first time in pre-B cells a link between CD9, RAC1 pathway and homing in testis which is a preferential site of late relapse. This work is supported by CNRS, University of Rennes 1, University Hospital of Rennes, Ligue Régionale contre le Cancer (sections 22, 35 and 56) (MPA, VG, MBT), Région Bretagne (MPA, MBT, VG), SFR Biosit UMS 3480 (VG, MBT), Association Laurette Fugain (VG). Disclosures No relevant conflicts of interest to declare.

9017 Background: Determination of the BRAF mutation status is a requirement for treatment selection of metastatic melanoma patients. However, the inter and intra-tumor genetic heterogeneity for BRAF status has been described supporting the need for testing multiple sites for the same patient. Circulating DNA is considered a promising non-invasive tool for cancer monitoring. The objective of our work was to determine the potential of plasma DNA to detect BRAF mutations, particularly in patients presenting discordant BRAF status from different tumor sites. Methods: Plasma DNA was extracted using the iPrep PureLink Virus Kit on an iPrep purification system (Invitrogen). BRAF mutations were detected using the Therascreen BRAF RGQ kit (Qiagen). Results: Ninety-nine patients were included, and plasma was collected at baseline. Nine stage III melanoma patients presenting a mutant BRAF (6 V600E, 2 V600K and 1 V600D) were tested and for 4 of them we detected the corresponding mutation (including the V600D and both...

9069 Background: Mutant BRAF DNA can be detected in plasma of melanoma patients bearing a mutation in their tumor. Detection of these genetic alterations in circulating DNA has potential clinical a...

Introduction CD9 is a membrane protein, member of the tetraspanin family. Recent publications have reported the role of CD9 on engraftment of hematopoietic stem cells, and on cancer stem cell potential. The expression of CD9 has been correlated to the risk of metastases and to a poor clinical outcome in various types of cancer. Surprisingly, CD9 protein is downregulated in ETV6/RUNX1 pre-B acute lymphoblastic leukemia. The purpose of our study is to investigate the effect of CD9 expression on migration and engraftment abilities of pre-B lymphocytes. Materials and Methods The CD9-positive Nalm6 and REH (ETV6/RUNX1) pre-B cells were used. By lentiviral transduction of shRNA targeting mRNA, we generated Nalm6 and REH cell lines depleted in CD9 protein. Engraftment tests were performed in vivo using Nod Scid Gamma immunodeficient mice. REH and Nalm6 cells were detected in bone marrow by CD10 and respectively CD19 or HLA-DR labelling. Ability of the different cell lines to adhere on fibronectin and to migrate through double chambers system in response to CXCL12 were measured in vitro. We also investigated the presence of membrane villosities on REH and REH shCD9 cell surface by scanning electron microscopy. Finally, F-actin polymerization after CXCL12 stimulation was measured by rhodamin-phalloidin labelling. Results In vivo engrafments tests showed that the human cells detected in bone marrow is strongly enriched in CD9 positive cells compared to the initially injected population. This result suggests that CD9 facilitates pre-B lymphobasts homing. An in vitro analysis of adhesion on fibronectin demonstrated that cellular adhesion is dependent on membrane expression of CD9. As well, the more CD9 is expressed, the higher the migration rate in response to CXCL12 chemokine is. The analysis of membrane villosities on REH cell surface revealed that cells over expressing CD9 had longer villosity than shCD9 transducted cell lines. Moreover, F-actin labelling after CXCL12 stimulation showed an increased F-actin polymerization in CD9-positive cells and the formation of actin extensions. Conclusion We provide novel evidence that CD9 is a key player of pre-B lymphoblasts engraftment, adhesion and CXCL12 dependant migration. CD9 expression is related to actin remodelling. We are now investigating a potential link between CD9 and RAC1 activation in response to CXC12. Therefore, the expression level of CD9 could impact leukemic blasts abilities to spread and be responsible of relapses. This work is supported by CNRS, University of Rennes 1, University Hospital of Rennes, la Ligue Régionale contre le Cancer (committee 22, 35 and 56) (MPA, VG, MBT), SFR Biosit UMS 3480 (VG, MBT), Association Laurette Fugain (VG) and Europe Career Integration Grant (MBT). Disclosures: No relevant conflicts of interest to declare.

e19151 Background: Detection of EGFR gene alterations is critical for predicting the response to tyrosine kinase inhibitors in patients with non-small-cell lung cancer. Analysis of other biomarkers might also be mandatory in the future, requiring more biological material. In routine practice, molecular testing is usually performed on tissues such as bronchial or transthoracic biopsies and surgical resections. We investigated the feasibility of using both cytology cell blocks and cell-free DNA from pleural effusions. Methods: Pleural effusion samples were centrifuged to form a pellet, fixed in formaldehyde, and processed as a cell block. DNA was extracted from 10-µm sections after paraffin removal using the Forensic kit and an iPrep system (Invitrogen). Cell-free DNA was extracted from the centrifugation supernatants using the QIAamp Circulating Nucleic Acid kit (Qiagen). Detection of molecular alterations were performed by allele-specific PCR (EGFR p.L858R mutation, BRAF p.V600E mutation), fragment size analysis (EGFR exon 19 deletions), and Sanger sequencing (KRAS mutations, codons 12 and 13). Results: Between January 2010 and December 2012, we tested 65 cell pellets (32 women and 33 men) and collected 34 paired supernatants. With the cell pellets, all the EGFR status were successfully determined (13 mutations; 20%), 9 mutations of KRAS on the 42 cases sequenced (17.6%), and 2 BRAF mutations on the 50 samples tested (4%) were found. We next investigated the potential use of the cell-free DNA from supernatants. Quantification by real-time PCR revealed significant amounts of DNA (mean = 306 ng extracted from 2 ml). EGFR, KRAS and BRAF tests were contributive for all these samples: 2 EGFR p.L858R mutations, 6 KRAS mutation (p.G12C, n=2; p.G12D, n=2; p.G13C, n=1 and p.G12F, n=1) and 1 BRAF p.V600E mutation were found. Most interestingly, the molecular status of the cell-free DNAs perfectly matched those obtained with the corresponding cell pellets. Conclusions: Pleural effusions provide high quality material for molecular testing and are suitable samples for routine practice. Cell-free DNA can also be successfully tested thus saving material for morphological / immunohistochemical / FISH analysis.

e19046 Background: Detection of EGFR alterations is critical for predicting the response to tyrosine kinase inhibitors (TKI) in patients with non-small-cell lung cancer (NSCLC). In clinical practice, molecular testing is performed on tumor tissues when available. We investigated the use of circulating cell-free DNA for the detection of EGFR alterations in patients with NSCLC. Methods: Serum and plasma were obtained by centrifugation (10 min, 2,000 g, 20°C) performed within 3 hours following blood collection. Cell-free DNA was extracted using the QIAamp Circulating Nucleic Acid kit (Qiagen). Detection of EGFR alterations was performed using the approved Therascreen EGFR RGQ kit (Qiagen). Samples were tested positive for EGFR mutation when the ΔCt value (Ct of the mutation specific PCR – Ct of the control PCR) was lower than 12 for exon 19 deletions, and below 14 for L858R mutation. Results: No EGFR alteration was detected in samples collected from healthy donors (n=6), NSCLC patients with a wild type EGFR (n=60), and early stages NSCLC patients presenting an EGFR mutation in their tumor (n=11). Thirteen metastatic patients presenting an EGFR mutation in their tumor were tested before initiation of TKI treatment. When available, both serum and plasma were analyzed. Nine of these patients (9/13; 69.2%) were tested positive in their serum. The ΔCt values obtained were lower for plasma than for serum in most cases, and more patients (10/11; 90.9%) were EGFR mutation positive when plasma was tested. Finally we tested 3 patients during TKI treatment on a monthly basis. For 2 patients we were unable to detect the mutation initially found in pretreatment samples. Clinically, both patients were partial responders. The third patient did not respond to TKI, and we detected the EGFR mutation with stable ΔCt values at all points tested. Conclusions: EGFR alterations can be found in patients presenting an EGFR mutation in a metastatic NSCLC. Plasma samples allowed a better detection rate. Our results suggest that DNA circulating in plasma is a useful alternative source of tumor DNA that could be used for determining EGFR mutation status, and for follow-up of treatment. Supported by a grant from Astra-Zeneca.

e21117 Background: Activating mutations in the tyrosine kinase domain of the EGFR gene in lung carcinoma has been associated with a dramatic response to tyrosine kinase inhibitors. Therefore, routine analysis of pathological specimens is mandatory in clinical practice to predict patient response. We have analyzed the results of the tests we performed during years 2010 and 2011. Methods: DNA was extracted from formalin-fixed paraffin-embedded tissues, following macrodisection, using an iPrep robot (Invitrogen). The p.L858R and p.L861Q substitutions (exon 21) were assessed by allele-specific PCR and exon 19 deletions by PCR amplification followed by DNA fragment analysis (polyacrylamide gel electrophoresis). These techniques allowed us to detect these mutations when present in at least 2% of the cells. Results: During these 2 years, we analyzed samples from 1,041 patients (including 907 adenocarcinomas, 71 large cell carcinomas and 28 squamous cell carcinomas). The EGFR mutational status could not be determined in 51 cases : DNA could not be amplified in 23 cases (2.2%), and 28 samples (2.7%) contained a limited percentage (