V. Scott

Outcome impact of PIK3CA mutations in HER2-positive breast cancer patients treated with trastuzumab

Background:Phosphatidylinositol 3-kinase (PI3K) pathway activation has been suggested to negatively influence response to anti-HER2 therapy in breast cancer patients. The present study focused on mutations of the PIK3CA gene, encoding one of the two PI3K subunits.Methods:PIK3CA mutations were assessed by direct sequencing in 80 HER2-positive patients treated with 1 year of trastuzumab. All patients preoperatively received four cycles of anthracycline-based chemotherapy, followed by four cycles of docetaxel and 1 year of trastuzumab, starting either before surgery with the first cycle of docetaxel and continuing after surgery (neoadjuvant trastuzumab arm, n=43), or only after surgery (adjuvant trastuzumab arm, n=37).Results:PIK3CA mutations were found in 17 tumours (21.3%). Better disease-free survival (DFS) was observed in patients with PIK3CA wild-type compared with mutated tumours (P=0.0063). By combining PIK3CA status and treatment arms, four separate prognostic groups with significantly different DFS (P=0.0013) were identified.Conclusion:These results confirm that the outcome of HER2-positive patients treated with trastuzumab is significantly worse in patients with PIK3CA-mutated compared with wild-type tumours.

Exonic expression profiling of breast cancer and benign lesions: a retrospective analysis

BACKGROUND Gene-expression arrays have generated molecular predictors of relapse and drug sensitivity in breast cancer. We aimed to identify exons differently expressed in malignant and benign breast lesions and to generate a molecular classifier for breast-cancer diagnosis. METHODS 165 breast samples were obtained by fine-needle aspiration. Complementary DNA was hybridised on splice array. A nearest centroid prediction rule was developed to classify lesions as malignant or benign on a training set, and its performance was assessed on an independent validation set. A two-way ANOVA model identified probe sets with differential expression in malignant and benign lesions while adjusting for scan dates. FINDINGS 120 breast cancers and 45 benign lesions were included in the study. A molecular classifier for breast-cancer diagnosis with 1228 probe sets was generated from the training set (n=94). This signature accurately classified all samples (100% accuracy, 95% CI 96-100%). In the validation set (n=71), the molecular predictor accurately classified 68 of 71 tumours (96%, 88-99%). When the 165 samples were taken into account, 37 858 exon probe sets (5.4%) and 3733 genes (7.0%) were differently expressed in malignant and benign lesions (threshold: adjusted p<0.05). Genes involved in spliceosome assembly were significantly overexpressed in malignant disease (permutation p=0.002). In the same population of 165 samples, 956 exon probe sets presented both higher intensity and higher splice index in breast cancer than in benign lesions, although located on unchanged genes. INTERPRETATION Many exons are differently expressed by breast cancer and benign lesions, and alternative transcripts contribute to the molecular characteristics of breast malignancy. Development of molecular classifiers for breast-cancer diagnosis with fine-needle aspiration should be possible.

Targeting the Deregulated Spliceosome Core Machinery in Cancer Cells Triggers mTOR Blockade and Autophagy

The spliceosome is a large ribonucleoprotein complex that guides pre-mRNA splicing in eukaryotic cells. Here, we determine whether the spliceosome could constitute an attractive therapeutic target in cancer. Analysis of gene expression arrays from lung, breast, and ovarian cancers datasets revealed that several genes encoding components of the core spliceosome composed of a heteroheptameric Sm complex were overexpressed in malignant disease as compared with benign lesions and could also define a subset of highly aggressive breast cancers. siRNA-mediated depletion of SmE (SNRPE) or SmD1 (SNRPD1) led to a marked reduction of cell viability in breast, lung, and melanoma cancer cell lines, whereas it had little effect on the survival of the nonmalignant MCF-10A breast epithelial cells. SNRPE or SNRPD1 depletion did not lead to apoptotic cell death but autophagy, another form of cell death. Indeed, induction of autophagy was revealed by cytoplasmic accumulation of autophagic vacuoles and by an increase in both LC3 (MAP1LC3A) protein conversion and the amount of acidic autophagic vacuoles. Knockdown of SNRPE dramatically decreased mTOR mRNA and protein levels and was accompanied by a deregulation of the mTOR pathway, which, in part, explains the SNRPE-dependent induction of autophagy. These findings provide a rational to develop new therapeutic agents targeting spliceosome core components in oncology.

Array CGH and PIK3CA/AKT1 mutations to drive patients to specific targeted agents: A clinical experience in 108 patients with metastatic breast cancer

Breast cancer includes high number of molecular entities targetable by specific agents. In this study, array CGH and PIK3CA/AKT1 mutations were used to drive patients into targeted therapy. A prospective molecular analysis was offered to metastatic breast cancer patients for whom samples were collected prospectively or retrospectively either from frozen or paraffin-embedded tissue. Analyses were performed using array CGH (Agilent platform) and PIK3CA (exon 10 and 21) and AKT1 mutations were explored by standard Sanger sequencing. One hundred and eight patients were included. Good quality CGH was obtained in 79% cases and was better for frozen samples. Genomic alterations were identified in 50% of patients including 11 PIK3CA and 8 AKT1 mutations. Eighteen treatments (17 patients) were administered according to their molecular profile with evidence of activity in nine. Reasons for not providing a genomic-driven treatment included absence of progressive disease (38%), investigators choice (9%), rapid PD (19%), and no drug access (21%). Array CGH correctly identified Her2 status in 97% cases; failures were related to low % of tumour cells. Our study showed that array CGH is feasible in the context of daily practice and, in combination with PIK3CA/AKT1 mutations, identifies a significant number of actionable molecular alterations that allow driving patients into specific targeted agents.

Abstract CT041: Anti-proliferative response and predictive biomarkers to palbociclib in early breast cancer: The Preoperative Palbociclib (POP) randomized trial

Background: In this monocentric randomized trial we aimed at identifying if short-term preoperative palbociclib treatment is associated with decreased proliferation and early biomarker changes in patients with early breast cancer (EBC) Methods: Untreated EBC patients were randomized 3:1 to oral palbociclib 125mg daily for 14 days until the day before the surgery vs no treatment. FFPE and frozen samples were extracted at baseline and at surgery. Primary objective was antiproliferative response defined as a natural logarithm of Ki67 expression at day 15 below 1. Immunostaining (Ki67, RB, pRB, p16, pAKT, pER), FISH (CCND1) and gene expression (GE) arrays were performed pre- and post-treatment. PIK3CA and AKT1 mutations were assessed pretreatment. Results: 74 patients were randomized in the palbociclib arm and 26 in the control arm. 93% tumors were HR-positive; 8% HER2-positive. Palbociclib treatment led to significantly more patients with antiproliferative response as compared to control (58% vs 10%, p = 0.0003). In addition, mean decrease from baseline in ln(Ki67) was significantly higher at day 15 in the palbociclib arm (p Citation Format: Monica Arnedos, Bianca Cheaib, Mohamed Amine Bayar, Stefan Michiels, Veronique Scott, Julien Adam, Valerie Leroux-Kozal, Virginie Marty, Chafika Mazouni, Benjamin Sarfati, Ivan Bieche, David Gentien, Suzette Delaloge, Magali Lacroix-Triki, Fabrice Andre. Anti-proliferative response and predictive biomarkers to palbociclib in early breast cancer: The Preoperative Palbociclib (POP) randomized trial. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr CT041.

BIRC5 (survivin): a pejorative prognostic marker in stage II/III breast cancer with no response to neoadjuvant chemotherapy.

PurposeNeoadjuvant systemic therapy (NAC) is currently used in the treatment of stage II/III breast cancer. Pathological complete response as a surrogate endpoint for clinical outcomes is not completely validated for all subgroups of breast cancers. Therefore, there is a need for reliable predictive tests of the most effective treatment.MethodsWe used a combination of predictive clinical, pathological, and gene expression-based markers of response to NAC in a prospective phase II multicentre randomized clinical trial in breast cancer patients, with a long follow-up (8 years). This study concerned the subpopulation of 188 patients with similar levels of pathological response rates to sequential epirubicin/cyclophosphamide and docetaxel to determine predictive marker of pCR and DFS. We used a set of 45 genes selected from high throughput analysis and a standardized RT-qPCR. We analyzed the predictive markers of pathological complete response (pCR) and DFS in the overall population and DFS the subpopulation of 159 patients with no pCR.ResultsIn the overall population, combining both clinical and genomic variables, large tumor size, low TFF1, and MYBL2 overexpression were significantly associated with pCR. T4 Stage, lymphovascular invasion, negative PR status, histological type, and high values of CCNB1 were associated with DFS. In the no pCR population, only lymphovascular invasion and high values of BIRC5 were associated with DFS.ConclusionsWe confirm the importance of ER-related and proliferation genes in the prediction of pCR in NAC-treated breast cancer patients. Furthermore, we identified BIRC5 (survivin) as a main pejorative prognostic factor in patients with breast cancers with no pCR. These results also open perspective for predictive markers of new targeted therapies.

Gene copy number variations in breast cancer of Sub-Saharan African women

The goal of this study was CGH array profiling of breast cancer from Malian women in order to define differences with those from USA. CGH array was performed in 28 samples, 17 with a triple negative phenotype. The profiles were compared to those of 106 tumors from USA. 6 chromosomal regions (6p21, 9q34, 11q13, 12q24, 17q25 and 22q12.1-22q13.1) were identified with a significant higher rate of copy number alterations. These regions contain several genes of interest including BCR. FISH and IHC confirmed that BCR was amplified and overexpressed particularly in triple negative tumors. Finally, 5 regions presented a high level of amplification in two or more samples, including 2 regions located between 9p22.3-9p23 and 9p23-9p24.1. This study confirms that breast cancers from African women present biological differences with those from USA. Larger studies are needed to go further in the identification of therapeutic targets that would be specific to African women.

801 ORAL High Throughput Molecular Analyses to Select Patients for Targeted Agents

Background: Rapid advancements in genomics paired with significant growth in the availability of targeted therapies offers clinicians expanding opportunities to provide increasingly effective cancer treatment. Currently, individual gene sequencing (e.g. EGFR) from formalin-fixed paraffinembedded (FFPE) tissue is widely used in cancer diagnosis. Shifting this paradigm towards NGS-based, comprehensive mutation testing in routinely collected FFPE cancer specimens will enable more complete and accurate characterization of patients’ cancers for individualized targeted therapy selection. Materials and Methods: DNA was extracted from 2×20micron sections of 83 specimens consisting of colorectal cancer, non-small cell lung cancer and melanoma. Hybridization-capture of 2574 exons across 176 oncogenes, tumour suppressor genes and ADME-related genes was performed to produce libraries appropriate for paired-end sequence analysis on the Illumina HiSeq2000 platform. Results: In-depth sequence analysis of 176 genes in 50 CRC, 29 NSCLC, and 4 melanoma specimens with median coverage averaging 213-fold (range 8 to 461) detected a per-sample average of 2 previously-described mutations, 7 novel mutations and 2 CNAs in the colon specimens, including frequent alterations in TP53 (33), APC (27), KRAS (12) and BRAF (6). The lung specimens averaged 1 previously described mutation, 8 novel mutations and 1 CNA per sample, most frequently KRAS (10), TP53 (7), JAK2 (3), EGFR (2) and BRAF (2). The melanoma cases exhibited on average 1 previously described mutation, 7 novel mutations and 3 CNAs including TP53 (4) and BRAF (2). In addition to validated clinically actionable mutations in EGFR, KRAS, and BRAF, and multiple alterations in well-known cancer genes such as TP53, STK11, APC,MLH1, BRCA2, and SMAD4, we detected many other mutations that are plausibly clinically actionable. These included activating mutations in the PI3 kinase subunit gene PIK3CA, as well as mutations in MET , KIT , ERBB2 and CDKN2A. Conclusions: It is feasible to perform highly sensitive and specific sequence analysis of hundreds of genes from routinely collected FFPE tissues. This approach detects not only the “hot spot” mutations commonly tested for in CRC, NSCLC and melanoma but also many additional mutations that could plausibly inform therapeutic decision-making. We suggest that clinical-grade next-generation sequencing should become a routine part of all clinical trials, and increasingly, of clinical care.

P5-01-07: Identification of SORBS2 as a Candidate Marker To Predict Metastatic Relapse in Breast Cancer.

Background Elucidation of promising cancer biomarkers from gene expression data can provide important insight into the relationship between signaling networks and cancer. SORBS2, sorbin and SH3 domain containing 2, is a multi-adapter protein involved in signal transduction associated to the cytoskeleton and was reported to be strongly repressed in pancreatic and cervical cancers. Methods: With the purpose of identifying genes involved in metastatic process, we compared gene expression profiling of 19 invasive ovarian cancers and 24 borderline tumors. Prognostic value of the selected genes was then tested in a gene expression array database that includes 1659 patients with early breast cancer (Gyorffy B et al. 2010). Upon isolation of SORBS2 as a predictor, its involvement in cell migration and tumor progression was investigated in vitro. Small interfering RNA targeting SORBS2 was used to downregulate its expression in T47D and Hela, two cell lines overexpressing SORBS2. Functional effect of siRNA-induced knockdown of SORBS2 on cell viability was determined by WST-1 assay and Trypan Blue exclusion test. Effect on cell migration was evaluated by wound-healing and transwell assays. Western blot analyses were also performed to examine the expressions of proteins involved in cell survival, death and migration. Results: High-throughput analyses of genes that are differentially expressed between borderline ovarian tumors and invasive carcinoma demonstrated that SORBS2 is significantly downregulated in invasive carcinoma (FDR Conclusion: This study is the first to provide evidence for an antiproliferative activity of SORBS2 with no effect on cell migration in breast cancer cells. Our clinical and in vitro data suggest that SORBS2 is a candidate marker to predict relapse in patients with early breast cancer. Molecular mechanisms mediating the antiproliferative effect of SORBS2 are currently being investigated. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-01-07.

High Resolution Oligonucleotide Array-CGH and Hot Spot Mutations To Select Patients for Targeted Therapies.

Background. A high number of genomic alterations (gene amplification, mutations) have been reported in breast cancers, and could potentially be targeted in early clinical trials. In the present molecular screening program, we aim at performing CGH array and hot spot mutations for 200 patients with metastatic breast cancer, in order to drive these patients in specific early clinical trials.Patients and Methods. Patients were selected to present a metastatic breast cancer under treatment in four French Centers. DNA was extracted by Qiagen method from either primary breast cancer or metastatic lesions. The hybridizations were done on 4*44K Agilent Array CGH. The analysis was done with Analytics v3.4.40 software. The file reported amplified zones (log2ratio>0.84 for biopsy, 1.58 for FNA) and focused on predefined genes that encode a candidate target. Hot spot mutations cover PI3K, AKT, PTEN genes. The primary endpoint was to determine how many patients were driven to a specific clinical trial because of molecular analyses. The secondary endpoints included PFS and response rates.Results. As june 12 th , 109 patients were included in the molecular screening program. CGH array was feasible and could be interpreted within one month in 85 (78%) patients. Array-CGH identified high level amplification in 57 patients (67%). As expected, the two most frequent amplicons included ERBB2 (17q11-12, 24%) and FGFR1 (8p11, 16%).The accuracy to detect ERBB2 amplification was 90%. A significant number of patients presented rare gene amplifications including ERBB4 amplification (n=1), CCND1/FGF4 amplification (n=6), FGFR2 (n=2), EGFR (n=1), VEGFB (n=4), BCR (n=1)Interestingly, only 4 patients presented TOP2A amplification. Hot spot mutations for PI3KCA, AKT, PTEN is being performedConclusion. This preliminary analysis shows that array CGH allows to identify patients who are candidate for targeted agents in phase I/II trials. Accrual will continue until 200 patients. Selected patients will be included in clinical trials in second semester of 2009 after a target validation by IHC. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 5067.