Search

Supplementary Data 1. This supplementary data comprises a step-by-step description on how our gene-expression based classification models were generated and validated

Supplementary Figure 1. Highlights the EFS and OS for the cohort of neuroblastoma patients with MYCN-amplified disease (n=114, Fig 1a) and for th subcohort of patients >18 months of age with stage 4, MYCN non-amplified disease (n=102, fig. 1b). F, favorable; UF, unfavorable.

Purpose: To optimize neuroblastoma treatment stratification, we aimed at developing a novel risk estimation system by integrating gene expression–based classification and established prognostic markers.Experimental Design: Gene expression profiles were generated from 709 neuroblastoma specimens using customized 4 × 44 K microarrays. Classification models were built using 75 tumors with contrasting courses of disease. Validation was performed in an independent test set (n = 634) by Kaplan–Meier estimates and Cox regression analyses.Results: The best-performing classifier predicted patient outcome with an accuracy of 0.95 (sensitivity, 0.93; specificity, 0.97) in the validation cohort. The highest potential clinical value of this predictor was observed for current low-risk patients [5-year event-free survival (EFS), 0.84 ± 0.02 vs. 0.29 ± 0.10; 5-year overall survival (OS), 0.99 ± 0.01 vs. 0.76 ± 0.11; both P < 0.001] and intermediate-risk patients (5-year EFS, 0.88 ± 0.06 vs. 0.41 ± 0.10; 5-year OS, 1.0 vs. 0.70 ± 0.09; both P < 0.001). In multivariate Cox regression models for low-risk/intermediate-risk patients, the classifier outperformed risk assessment of the current German trial NB2004 [EFS: hazard ratio (HR), 5.07; 95% confidence interval (CI), 3.20–8.02; OS: HR, 25.54; 95% CI, 8.40–77.66; both P < 0.001]. On the basis of these findings, we propose to integrate the classifier into a revised risk stratification system for low-risk/intermediate-risk patients. According to this system, we identified novel subgroups with poor outcome (5-year EFS, 0.19 ± 0.08; 5-year OS, 0.59 ± 0.1), for whom we propose intensified treatment, and with beneficial outcome (5-year EFS, 0.87 ± 0.05; 5-year OS, 1.0), who may benefit from treatment de-escalation.Conclusions: Combination of gene expression–based classification and established prognostic markers improves risk estimation of patients with low-risk/intermediate-risk neuroblastoma. We propose to implement our revised treatment stratification system in a prospective clinical trial. Clin Cancer Res; 21(8); 1904–15. ©2014 AACR.See related commentary by Attiyeh and Maris, p. 1782

Supplementary Table 1. Clinical co-variates for the 709 patients who participated in the study. This supplementary table comprises detailed information on clinical co-variates for all 709 neuroblastoma patients who participated in the study.

Supplementary Table 2. External performance validation of the top five classifiers. This supplementary table comprises the external performance metrics of the top five classifiers. Indicated are the values for classification accuracy, sensitivity , specificity and Matthew's correlation coefficient (MCC) in the prediction of 325 patients of the test set who fulfilled the criteria for classifier training (Favorable, n=187; Unfavorable, n=138).

Supplementary Table 4. Transcripts that contribute to the SVM_th10 classifier. This supplementary table highlights the transcripts that constitute the SVM_th10 classifier. Indicated for each feature are the Oligo-ID, the Gene symbols, the Entrez gene IDs, the RefSeq IDs, the Gene IDs and the Transcript IDs.

Legends to Supplementaries. This documents comprises the legends to the supplementary figures and tables

Supplementary Data 2. This supplementary data comprises the r-algorithm scripts required to perform the both model selection and validation as described in the step-by-step protocol

Background: The status of the HER2 gene is not static and may change between the primary tumor, lymph node metastases and distant metastases. This status change can be a consequence of the natural evolution of the tumor or can be induced by therapy. The HER2 gene status is, in the majority of cases, established at the moment of diagnosis. After chemotherapy, monitoring this marker can be a real challenge because of ploidy changes induced by drugs. We made a HER2 genetic examination by Fluorescent in situ Hybridization (FISH) of invasive breast cancers before and after taxane treatment. After treatment we found supernumerary HER2 gene copies up to 15 copies in the cases non-amplified at diagnosis. Material and methods: All 410 invasive breast carcinoma samples were collected in a single laboratory (Service de Pathologie, Centre Jean Perrin). Tumor samples were collected by biopsy at diagnosis or by surgical excision after taxanes chemotherapy. For all cases we had matched samples, one before and one after therapy. Tumor samples were formalin-fixed paraffin-embedded. IHC staining was done to detect the expression of HER2 protein. Only cases which were IHC 2+ or IHC 1+ (with scattered HER2 heterogeneous staining) were checked by FISH. Results: After taxane-based neoadjuvant chemotherapy (4–8 cycles) thirty patients (HER2 negative at Dg) out of 344 showed supernumerary HER2 gene copies detected in giant cells with large polyploid nuclei. The proportion of polyploid post-therapy cells among regular-size cancer cells (diploid) was from 15 to 30%. The FISH examination of polyploid cells revealed an increase of HER2 copy number (and chromosome 17centromere) up to 15. The same copy number raise was observed with EGFR and ALK FISH probes. The ratio of all 30 cases with two distinct tumor cell populations (near diploid cells and giant cells) was up to 1.65 (not amplified). In one third of cases the average number of HER2 gene copies was higher than 6. Conclusions and recommendations: As a result of taxane administration, polyploid cells are formed and that most (if not all) chromosomes have been replicated simultaneously because of absent chromosomal segregation secondary to microtubule inhibition by taxanes. This HER2 copy increase is distinct from the mechanisms accounting for a more restricted gene amplification during carcinogenesis. The viability of giant polyploid cells is particularly low and most if not all of these cells succumb to mitotic catastrophe, an onco-suppressive mechanism causing cell cycle arrest and cell death. Irrespective of this consideration, it would be a critical error to classify cases as HER2 amplified solely only the bases of HER2 copy number >6. We recommend that the detection of more than 6 HER2 copies after taxane-based chemotherapy should be prompt a validation step involving the exclusion of polyploid cells from the analysis, as well as hybridization with a second and third probe specific for a chromosome other than 17. This validation step appears crucial to avoid unnecessary, costly treatments with HER2-targeted therapies after the failure of taxane-based chemotherapies. The clinico-pathological parameters of tumors with giant cells will be discussed. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-08-04.

Background The genetic heterogeneity of HER2 gene amplification (GA) in breast cancer has previously been described, but the clinical significance of this phenomenon remains unknown. We studied the genetic categories of a series of consecutive 2+ IHC cases over 5 years, with a focus on cases with HER2 GA detected in minor clone(s). We compared the HER2 status in primary tumors and positive axillary lymph nodes (ALN) when available to test the hypothesis that HER2 amplified cells are more aggressive and metastase quicker than non amplified cells. Material and methods: From January 1st. 2006 to May 30. 2011, 4491 invasive breast carcinomas had HER2 immunohistochemical (IHC) and/or HER2 gene status evaluation on their tumor sample in Institut Gustave Roussy. The distribution according to their IHC was as follows: 0 in 2915 cases (65%), 1+ in 569 cases (12.6%), 2+ in 536 cases (11.8%) and 3+ in 471 cases (10.6%). All 2+ samples were checked by Fluorescence in situ hybridization (FISH). For each case we analysed 100 invasive cells in 10 microscopic fields. Heterogeneous amplification was defined as presence of 5–50% of amplified cells (ratio>2,2) within the tumors otherwise classified as not amplified or borderline (ratio for 100 cells: Results: FISH with HER2/cen17 probes can classify IHC 2+ tumors into 5 genetic categories: normal gene status (two chromosomes (chr) 17 and two HER2 genes): 90 cases; chr 17 “ polysomy “/17q gain: 145 cases; HER2 gain: 122 cases; HER2 amplification (major clone): 135 cases and chr 17 monosomy: 44 cases. We focused our interest to IHC 2+ cases where the major clone showed no HER2 amplification, but minority clone (cut off 5%) showed the HER2 amplification (ratio >2,2 or >6 HER2 copies per cell). We found 48 such cases (10% of 2+ cases). These cases had been reported as not amplified (42/48) or borderline (6/48). To find out if the amplified cells are those which metastase, we checked 10 involved ALN: they all showed the similar genetic heterogeneity as a primary tumor with a global ratio < 2,2. Other cases with involved ALN are in ongoing study by FISH. Conclusions: The genetic heterogeneity of IHC 2+ tumors is a common event and five different genetic categories can be detected. A part of the HER2 negative cases contained one or several HER2 amplified clones. Preliminary results show a similar genetic heterogeneity of metastatic population. Our results on the small series of cases do not confirm the hypothesis that the only amplified clones metastase, but all clones (amplified and not amplified) has a metastatic capacity. A multi-center collaboration is needed to collect the high number of cases with heterogeneous amplification to: 1) Find out a proportion of such patients in HER2 IHC 2+ group (10% in our study). 2) Study a status of the gene in relapse cases. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-05-07.

Purpose: CTCs represent an alternative source of tumor material potentially usable as a surrogate tissue marker in the context of targeted treatments. The objective of the present study was to analyze the status of key molecular markers such as ERG gene rearrangement and AR amplification in tumor tissues (primitive tumor specimens and metastasis biopsies) and CTCs of patients with metastatic prostate cancer (PCa). Patients and Methods: CTCs were detected using two methods, CellSearch (Veridex LLC, Raritan, NJ, USA) and filtration (ISET, isolation by Size of Epithelial Tumor cells) combined to four-color immunofluorescent staining. Epithelial (cytokeratins, EpCAM (Epithelial Cell Adhesion Molecule)) and mesenchymal (vimentin, N-cadherin) markers were tested. Molecular characterization of CTCs was performed on filters by a two step method combining four color immunofluorescent staining and Filter Adapted-Fluorescent In situ Hybridization (FA-FISH). Image analysis was performed using an ARIOL automated scanner (LEICA) to precisely relocate FISH signals in phenotypically different CTC subsets. ERG gene rearrangement and AR copy number gains were examined both in CTCs enriched on filters and in the cell fractions recovered from the magnests after CellSearch and spread on slides for FISH. Primitive tumor specimens and metastasis biopsies were also collected and analyzed by FISH. Results: Higher levels of CTCs were detected by filtration enrichment combined to four-color immunofluorescent staining compared to CellSearch. Filtration enriched CTCs expressed epithelial, mesenchymal and hybrid (both epithelial and mesenchymal) markers suggesting the existence of CTCs at different stages of the epithelial to mesenchymal transition (EMT) process in most PCa patients. A higher level of gene variability and tumor heterogeneity was observed in filtration enriched CTCs compared to those captured by the CellSearch. ERG gene rearrangement and AR amplification in prostate tissue, metastases, and filtration enriched CTCs were analysed by FISH. ERG gene rearrangement and AR copy number gain status present within the tumor tissue was in concordance with that found in filtration enriched CTCs. However, FISH experiments also revealed a much higher degree of cell-to-cell variability and tumor heterogeneity in filtration enriched CTCs compared to tumor tissues. Conclusion: We show here that prostate CTCs are highly heterogeneous in term of EMT markers and ERG gene rearrangement and AR amplification marker status. The high degree of cell-to-cell variability in ERG and AR gene patterns present in CTCs reveals important tumor heterogeneity and could have important implications for developing personalized anticancer strategies based on the use of CTCs in PCa patients. Citation Format: Christophe Massard, Marianne Oulhen, Alexander Valent, Sylvestre Lemoulec, Karim Fizazi, Philippe Vielh, Françoise Farace. Characterization of the molecular heterogeneity of circulating tumor cells in metastatic prostate cancer. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3079. doi:10.1158/1538-7445.AM2014-3079

Background: Hormonal therapy is considered one of the most effective treatment options for patients with estrogen receptor (ER) and/or progesterone receptor (PR) positive breast cancers (BC); however only a 50% objective response rate is observed as 1st line therapy of ER positive (ERpos) tumors (Bonneterre, 2001). An IHC diagnostic test to identify the activated form of the ER and PR has previous been described (ASCO 2013 abst # 592, 593 & 5602). The nuclear morphological of distribution of the ER and PR can be categorized in two patterns, a presumably not-activated D (diffuse pattern) or activated A (aggregated pattern). This classification is correlated with antiestrogen treatment outcome (ASCO 2013, abst # 592), and correlated with antiprogestin activity in vitro (AACR 2013, abst # 1317). Clinical activity of antiprogestins has been previously described (Robertson 1999, Jonat 2002). It is generally accepted that expression of PR is under the transcriptional control of ER, thus targeting ER can potentially suppresses PR expression. However, in breast cancer tissues although both ER and PR are present in the tumor specimen, there is low level of co-expression of ER/PR at the cellular level (ASCO 2013, abstract # 596). The goal of the study is to elucidate whether or not the activated form of ER and PR were coexpressed. Method: 293 ERpos or PRpos BC were obtained from the Oscar-Lambret Cancer Center, Lille, France. Each sample was analyzed with specific antibodies for ERα, PR A or PR B; a subset (53 for PR A and 49 for PR B cases) was analyzed by immunochemical duel staining on the same paraffin tumor section for ERα and PR A or PR B. Tumors were deemed receptor positive if >10% of the tumors cells (tc) were stained. Samples processed for duel staining came from the same dataset and were selected if they had both ERα and PR A and/or PR B present in more than 10% and less than 80% of stained tc. The distribution of ERα, PR A, PR B tc was described. Determination of the activated form of ERα (AERpos) and PR A (APRpos A), PR B (APRpos B) was performed using the previously described IHC technique at 100X magnification. Results: Standard receptor positivity (% of tumors tested) was as follows; ERα 85%, PR A 61% and PR B 65%, either PR A or B 68%. Duel staining receptor positivity (% of tumors) showed for the PR A subset: 72% ER pos, 74% PR A pos, and 9% ER/PR (% of tumors co-expressing); for the PR B subset: ER 60%, PR B 100%, and 13% ER/PR (% of co-expressing); Subnuclear Morphology: 216 cases were ER and PR positive, and of these, 23% were AERpos, 23% were APRpos A, 24%, APRpos B% and 31% APRpos for either A or B. In the 216 case dataset, 18% of the tumors had both AER and APR A or APR B; However, using the duel staining method no cases had more than 10% of the tc co-expressing i.e. AERpos and APRpos, indicating that the co-expression was principally at the tumor level and not at the cellular level. Conclusion: As known, BC commonly express both ER and PR; however the co-expression of ER and PR in malignant cells is uncommon, and this co-expression of ER and PR could be explained by chance. Although AER and APR are expressed within the same tumor, the activated forms of the ER and PR are not co-expressed at the cellular level. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P6-05-13.

Purpose: The diagnostic test of ALK rearrangement in non-small-cell lung cancer (NSCLC) for crizotinib treatment is currently done on tumor biopsies or fine needle aspirations. The present study was designed to evaluate -1) whether ALK-rearrangement diagnosis could be performed using circulating tumor cells (CTCs), -2) whether CTCs harboring ALK-rearrangement could be monitored in ALK-positive patients treated by crizotinib. Patients and Methods: CTCs were isolated in 18 ALK-positive and 14 ALK-negative patients by blood filtration using ISET (Isolation by Size Epithelial Tumor cells) and tested by filter adapted-fluorescence in situ hybridization (FA-FISH), a FISH method optimized for filters. Numbers of ALK-rearranged cells and patterns of ALK-rearrangement were determined in CTCs and compared to those present in tumor biopsies. ALK-rearranged CTCs and tumor specimens were characterized for epithelial (cytokeratin, E-cadherin) and mesenchymal (vimentin, N-cadherin) markers expression. ALK-rearranged CTCs were monitored in ALK-positive patients treated by crizotinib. Results: ALK-rearranged CTCs [ranging from 4 to 34 CTCs /1mL] were detected in all ALK-positive patients, while no or only one ALK-rearranged CTCs was detected in blood samples obtained from ALK-negative patients. ALK-rearranged CTCs harboured a unique (3’ 5’) split pattern while heterogeneous patterns (3’ 5’, only 3’) of splits were present in tumors. ALK-rearranged CTCs exclusively expressed a mesenchymal phenotype and contrasted with heterogeneous epithelial and mesenchymal marker expressions in tumors. Variations in levels of ALK-rearranged CTCs and in CTCs harboring a gain of ALK native copies were detected under crizotinib treatment. Conclusion: We report that ALK-rearrangement can be detected in CTCs of ALK-positive NSCLC patients using ISET and FA-FISH, enabling diagnostic testing for crizotinib treatment. Our results suggest that CTCs harboring a unique ALK-rearrangement and mesenchymal phenotype may arise from the clonal selection of tumor cells that have acquired the potential to drive metastatic progression of ALK-positive NSCLC. The molecular characterization of CTCs in patients undergoing crizotinib treatment might provide new insights into mechanism of resistance to ALK tyrosine kinase inhibitors, and possible strategies to overcome this resistance. Citation Format: Emma Pailler, Julien Adam, Amélie Barthelemy, Marianne Oulhen, Nathalie Auger, Alexander Valent, David Planchard, Melissa Taylor, Isabelle Borget, Fabrice Andre, Jean-Charles Soria, Philippe Vielh, Benjamin Besse, Françoise Farace. Detection of circulating tumor cells harboring a unique ALK rearrangement in ALK-positive non-small cell lung cancer. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1470. doi:10.1158/1538-7445.AM2013-1470

Purpose: CTCs represent an alternative source of tumor material potentially usable as a surrogate tissue marker in the context of targeted treatments. The objective of the present study was to characterize the status of key molecular markers such as ERG gene rearrangement, and AR and c-myc amplifications, according to epithelial and/or mesenchymal marker expression in CTCs of patients with metastatic prostate cancer (PCa). Patients and Methods: CTCs were isolated from the blood of 40 patients at day 0 and day 30 of abiraterone treatment. CTCs were detected using two methods, CellSearch® (Veridex LLC, Raritan, NJ, USA) and ISET (Isolation by Size of Epithelial Tumor cells). Molecular characterization of CTCs was performed on ISET filters by combining four color immunofluorescent staining and Filter Adapted-Fluorescent In situ Hybridization (FA-FISH®). To optimize pan-keratin detection, four pancytokeratin antibodies, including AE1/AE3, KL1, C11, and A45-B/B3, were tested on the LnCAP prostate cell line spiked in normal blood cells and on PCa CTCs enriched by ISET. Epithelial (cytokeratins, E-cadherin, EpCAM (Epithelial Cell Adhesion Molecule)), mesenchymal (vimentin, N-cadherin), PTEN, androgen receptor (AR) and Ki67 marker expression was detected by immunofluorescent staining on ISET filters according to a method we previously established. FA-FISH® was optimized for the detection of ERG gene rearrangement and AR and c-myc amplification or gain of copy number. A two step method combining immunofluorescent staining and FA-FISH® was established on an ARIOL automated scanner (LEICA) to precisely relocate FISH signals in phenotypically different CTC subsets. Results: Higher CTC levels were detected by ISET compared to CellSearch®. Quantitative analysis of the intensity values for each tested pan-keratins antibodies showed that KL1 and A45-B/B3 clones gave the strongest immunofluorescent signals both on the LnCAP cell line and PCa CTCs. CTCs expressing epithelial, mesenchymal and hybrid (both epithelial and mesenchymal) markers were detected in most patients suggesting the existence of CTCs at different stages of the epithelial to mesenchymal transition (EMT) process. The loss of PTEN expression was also observed using hematopoietic cells present on ISET filters as a positive control of PTEN expression. ERG gene rearrangement, AR and c-myc amplifications were detected. The presence of these markers within CTC subsets expressing epithelial and/or mesenchymal markers is currently studied. Conclusion: Several molecular markers such as ERG gene rearrangement, AR and c-myc amplifications were detected in PCa CTCs. The presence of these markers in CTCs according to their EMT status will be presented. Citation Format: Marianne Oulhen, Christophe Massard, Alexander Valent, Rachel Young, Sylvestre Le Moulec, Karim Fizazi, Philippe Vielh, Françoise Farace. Characterization of the molecular heterogeneity of circulating tumour cells in metastatic prostate cancer. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1471. doi:10.1158/1538-7445.AM2013-1471

Background: Antihormonal agents (AH) are some of the most clinically useful anticancer drugs. AH resistance and the availability of alternative treatments have made patient selection increasingly important for AH drug development. We hypothesized that the progesterone receptor (PR) may be constitutively active because of the aberrant biological pathways and that selecting patients with baseline tumor PR activation would predict for antiprogestin activity. Activated PR (APR) can be visualized in nuclei as fluorescent aggregates via fluorescent green protein engineered PR cell lines in vitro and by immunofluorescence (IF) in tissues. The detection of PR foci has been associated with transcription and gene activation in vitro. PR phosphorylation (phos) is induced by a variety of pathways. We have developed a IHC method that allows APR determination in tumor biopsies on a routine basis (SABCS 2012). The predictive nature of the observation of APR by IHC in vitro has been tested with onapristone (ONA) for cell survival and proliferation. Methods: 8 human breast cancer (BT-474, CAMA-1, EVSA-T, HCC-1954, MCF-7, MDA-MB-231, T-47D, ZR-75-1) and 2 human endometrial cancer (Ishikawa, HEC-1-A) cell lines were studied. Cytoblocks for APR analysis were made for each time point and experimental condition, baseline ER/PR expression was determined by IHC and PR phos was analyzed with specific antibodies (pSER162, 190, 294, 400, 554) by Western Blotting and IHC. Culture conditions: normal and stripped FBS, +/- ONA and growth factor/hormone exposure (EGF, FGF2, E2 and P4) in triplicate. Analysis was performed at baseline (BL), 6h, 96h and 7 days. Cell viability was assessed by MTS assay, proliferation was studied by Ki67 analysis. APR foci were determined by IHC and read by a pathologist blinded to the experimental conditions. Results: At BL only two cell lines were PR pos/APR pos (T47D, CAMA-1) and all other cell lines where PR pos/APR neg (BT-474, EVSA-T, HCC-1954, MCF-7) or PR neg/APR neg (Ishikawa, HEC-1-A, MDA-MB-231, ZR-75-1). ONA exerted inhibitory effect only in APR pos cell lines. In stripped FBS at 6h, T47D APR pos status was converted to APR neg by ONA except with E2 stimulation which required longer treatment. The CAMA-1 data are not yet available. No APR neg cell lines were converted into APR pos except in one experiment after protracted exposure to EGF. ONA generally decreased phos; phos decrease measured by WB did not correlate with growth inhibition. Phos status determined by IHC and Ki67 proliferation assessment are currently in process and will be presented with the confirmatory data. Conclusions: Baseline APR positivity is predictive of ONA activity in vitro. ONA converts APR pos cells to APR neg within 6h in most cases, a time frame consistent with PR biology. The growth inhibitory effects of ONA were not correlated to PR phos in the cell lines and conditions tested. The APR IHC methodology may be applicable in the clinical setting. Citation Format: Erard Gilles, Laura Caplier, Alexander Valent, Guillaume Serin, Anne Gompel, Jacques Bosq, Alexander Zukiwski. Antiprogestin drug development: in vitro validation of a potential clinical biomarker. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1317. doi:10.1158/1538-7445.AM2013-1317

Introduction: ERCC1 allows removal of branched DNA structures at essential steps in DNA repair pathways such as Nucleotide Excision Repair (NER), Fanconia-dependent Interstrand Cross-Link Repair (ICL-R), long-patch base excision repair (BER), and Single Strand Annealing (SSA). ERCC1 expression is a prognostic and predictive biomarker of chemotherapy effect in NSCLC. The most frequent method to determine ERCC1 status is the evaluation of gene expression (IHC or RT-PCR). Here, we studied the prevalence of ERCC1 copy number variation (CNV) using FISH. Methods: FISH was performed with Poseidon™ dual color probe set, ERCC1 (19q13) and ZNF443 (19p13) on a panel of 17 NSCLC cell lines and 324 FFPE tumor samples from resected NSCLC patients (103 from IGR, Villejuif, France, 221 from MDACC, Houston, USA). Gene amplification (GA) and high polysomy (HP) were defined following actual consensus rules for EGFR assessment. Results: ERCC1 copy number was increased (>2) in a large majority of cell lines (13/17) of which three displayed gene amplification (GA) (H2342, CN=5; H1299, CN>12; HOP62, CN>15). In human samples, overall ERCC1 gene abnormalities were detected in 65 cases (20.1%) including 31 GA (6.9%, the majority with clusters) and 34 HP (13.7%, more frequent in squamous cell carcinomas). In patients with ERCC1 expression status available (immunohistochemistry) we found a non-significant trend for higher ERCC1 expression in HP patients compared to GA and low CNV patients (P=0.06). Despite a 3-fold increase in IC50 of cisplatin in cell lines with ERCC1 gene amplification compared to normal cells (with no CNV) the ERCC1 FISH status seemed unable to predict chemosensitivity in resected NSCLC patients. In patients not treated with chemotherapy, DFS was significantly longer in patients with normal ERCC1 FISH status of the tumor compared to patients with ERCC1 gene amplification (P=0.02). Conclusion: ERCC1 copy number variation is a frequent event in NSCLC and could reflect genomic instability, but the clinical relevance of this observation needs further investigation. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 3098. doi:1538-7445.AM2012-3098

Introduction: The anaplastic lymphoma kinase (ALK) gene translocation is a rare event, described in approximately 4-6% of lung adenocarcinoma. ALK translocation is a robust predictive factor for ALK inhibitors sensitivity. However, increased ALK gene copy number seems to be a frequent event, described in 13-17% of Non-small cell lung cancer (NSCLC). The goal of this study was to explore the predictive value of ALK High copy number for 2 ALK inhibitors, crizotinib and TAE684 in NSCLC cell lines. Methods: 27 NSCLC cell lines were screened for ALK copy number by FISH after cytospin. ALK high copy number (HCN) was defined as the presence of ≤ 5 copies of ALK. FISH with CEP2 was performed to determine the ploïdy status in cell lines with HCN. As crizotinib is a dual c-MET and ALK inhibitor, FISH c-MET was performed in selected cell lines for in vitro studies. C-MET scoring was based on Cappuzzo criteria. In vitro sensitivity was evaluated through WST1 assays and clonogenic tests. Kelly, an ALK mutated neuroblastic cell line, was added as a positive control. Cells were seeded in triplicates in 96-well plates 18h before the treatment with TAE 684 or crizotinib with concentrations varying from 50 to 4000nM and 100 to 10000nM respectively. The half maximal inhibitory concentration of the two drugs (IC50) was estimated after 72h of treatment with 4 and 5 parameters logistic regression models. In clonogenic tests, crizotinib was tested with three schedules of concentrations: low dose (50nM), intermediate (200nM) and high (1000nM). Results: Among NSCLC cell lines, 6 (22%) displayed more than 5 copies of ALK, 19 (70%) presented a gain of 3 or 4 ALK copy number, only one cell line exhibited normal ALK copies and one harbored EML4-ALK translocation. FISH with CEP2 revealed a polysomy of chromosome 2 in cases with ALK HCN. C-MET amplification was observed in only three cell lines. Based on ALK FISH status, 9 NSCLC cell lines were selected to undergo in vitro assays, 6 exhibiting ALK HCN (H2030, H661, A427, H3255, BEN, H1299) and 3 exhibiting low ALK copy number (LCN) (H1975, H1651, H1650). Kelly and A549 cell lines were added as sensitive and resistant controls respectively. In ALK HCN cell lines, two (H661 and A427) were as sensitive as Kelly for crizotinib through clonogenic and WST1 assays. In addition, ALK HCN cell lines exhibited lower IC50 for anti-ALK drugs than cell lines with ALK LCN (median IC50 with crizotinib values: 1750nM [300-2800nM] in ALK HCN cell lines versus 4500nM [800-8000nM] in ALK LCN cell lines, p=0.35). The trend of crizotinib activity was not due to c-MET inhibition. Conclusion: The in vitro assays suggest that ALK HCN may be a predictive marker for sensitivity to crizotinib. The predictive value of ALK HCN would be investigated in in vivo models. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1715. doi:1538-7445.AM2012-1715