Abstract 3697: SIRT2 is a modulator of response to targeted therapies through regulation of MEK kinase activity

Resistance to targeted therapies is a major problem in cancer treatment. The Epidermal Growth Factor Receptor (EGFR) antibody drugs are effective in a subset of colorectal cancers, but the molecular mechanisms of resistance are understood poorly. Genes involved in epigenetic regulation are frequently deregulated in cancer, raising the possibility that such genes also contribute to drug resistance. To address this question, we compiled an shRNA library for chromatin modifying enzymes, chromatin remodelers, and many other factors associated with chromatin regulation. Using this library, we performed a loss of function genetic screen in colon cancer cell lines sensitive to EGFR inhibitors. We identified SIRT2, a NAD+ dependent deacetylase, as a modulator of the response to EGFR inhibitors in colon and lung cancer. Mechanistically, loss of SIRT2 expression enhanced MEK acetylation, resulting in increased ERK activity. Conversely, overexpression of SIRT2 led to reduced phospho ERK and inhibition of growth. SIRT2 loss also conferred resistance to BRAF and MEK inhibitors in BRAF mutant melanoma and KRAS mutant colon cancers, respectively. These results warrant a further investigation into a potential role of SIRT2 in resistance to drugs that act in the Receptor Tyrosine Kinase-RAS-RAF-MEK-ERK signaling pathway.

Citation Format: Prashanth K. Bajpe, Anirudh Prahallad, Hugo Horlings, Iris Nagtegaal, Roderick Beijersbergen, Rene Bernards. SIRT2 is a modulator of response to targeted therapies through regulation of MEK kinase activity. [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 3697. doi:10.1158/1538-7445.AM2014-3697

Abstract 2767: Identification of kinase fusion genes in bladder cancer through kinome-centered RNA sequencing

Urothelial carcinoma (UC) of the bladder is one of the most common cancers worldwide. The treatment of patients with metastatic or locally advanced UC consists of platinum-based chemotherapy. Unfortunately, a large number of patients will ultimately develop resistance. Therapeutic options at that point are very limited: the last FDA drug approval for the treatment of bladder cancer dates back more than 2 decades. Identification of activated signaling pathways in UC can provide new targets for treatment. Recent DNA and RNA sequencing projects in invasive UC have revealed somatic mutations in several cancer genes. Some of these mutated genes, such as FGFR3 and PIK3CA, could potentially guide therapy. However, our knowledge of gene rearrangements in bladder cancer remains limited. These fusion genes often involve multiple fusion partners, as was reported for FGFR3 fusions in bladder cancer, which represents a significant challenge for discovery and for subsequent diagnostic screening. Therefore, a global detection method is needed to fully understand the diversity of alterations driving this disease. We used a high-throughput platform to systematically profile kinase fusions through specific enrichment of kinase transcripts. Using this approach, we screened 80 muscle invasive UC specimens and identified a number of activating mutations and novel fusion transcripts. The fusion genes identified in this discovery set will be validated in a second cohort of UC specimens to determine their frequency. Functional validation will be presented for some of these fusion genes. These genetic alterations may provide new avenues for individualized molecular treatment of UC patients.

Citation Format: Floris Groenendijk, Iris de Rink, Laura Mertens, Annegien Broeks, Yann Neuzillet, Jeroen de Jong, Bas van Rhijn, Rene Bernards, Michiel van der Heijden. Identification of kinase fusion genes in bladder cancer through kinome-centered RNA sequencing. [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 2767. doi:10.1158/1538-7445.AM2014-2767

Hurdles on the road to personalized medicine

Cancer treatment is slowly shifting from an approach in which the tissue of origin and the histology were the guiding principles for the choice of chemotherapy towards a genotype-centric approach in which the changes in the cancer genome are used to select patients for treatment with highly selective and targeted drugs. This transition has all the hallmarks of a disruptive innovation and requires major adjustments in the way that cancer is diagnosed and treated. We discuss here the hurdles on the road ahead to a more personalized treatment of cancer.

The Good, the Bad, and the Ugly: in search of gold standards for assessing functional genetic screen quality

Variable screen quality, off-target effects, and unclear false discovery rates often hamper large-scale functional genomic screens in mammalian cells. Hart et al (2014) introduce gold standard reference sets of essential and non-essential genes, aiming at standardizing the analysis of genome-wide screens. This work provides a framework to compare both the quality and analysis methods of functional genetic screens.

The NuRD complex cooperates with DNMTs to maintain silencing of key colorectal tumor suppressor genes

Many tumor suppressor genes (TSGs) are silenced through synergistic layers of epigenetic regulation including abnormal DNA hypermethylation of promoter CpG islands, repressive chromatin modifications and enhanced nucleosome deposition over transcription start sites. The protein complexes responsible for silencing of many of such TSGs remain to be identified. Our previous work demonstrated that multiple silenced TSGs in colorectal cancer cells can be partially reactivated by DNA demethylation in cells disrupted for the DNA methyltransferases 1 and 3B (DNMT1 and 3B) or by DNMT inhibitors (DNMTi). Herein, we used proteomic and functional genetic approaches to identify additional proteins that cooperate with DNMTs in silencing these key silenced TSGs in colon cancer cells. We discovered that DNMTs and the core components of the NuRD (Mi-2/nucleosome remodeling and deacetylase) nucleosome remodeling complex, chromo domain helicase DNA-binding protein 4 (CHD4) and histone deacetylase 1 (HDAC1) occupy the promoters of several of these hypermethylated TSGs and physically and functionally interact to maintain their silencing. Consistent with this, we find an inverse relationship between expression of HDAC1 and 2 and these TSGs in a large panel of primary colorectal tumors. We demonstrate that DNMTs and NuRD cooperate to maintain the silencing of several negative regulators of the WNT and other signaling pathways. We find that depletion of CHD4 is synergistic with DNMT inhibition in reducing the viability of colon cancer cells in correlation with reactivation of TSGs, suggesting that their combined inhibition may be beneficial for the treatment of colon cancer. Since CHD4 has ATPase activity, our data identify CHD4 as a potentially novel drug target in cancer.

Unlikely suspects identified in neuroblastoma conspiracy

KIF1B is a candidate tumor-suppressor gene in neuroblastoma whose function is to mediate apoptosis when nerve growth factor becomes limiting in the developing nervous system. Chen and colleagues now provide mechanistic insight into how one of the protein products of this locus, KIF1Bβ, induces apoptosis.

Reversible and adaptive resistance to BRAF(V600E) inhibition in melanoma

Treatment of BRAF(V600E) mutant melanoma by small molecule drugs that target the BRAF or MEK kinases can be effective, but resistance develops invariably. In contrast, colon cancers that harbour the same BRAF(V600E) mutation are intrinsically resistant to BRAF inhibitors, due to feedback activation of the epidermal growth factor receptor (EGFR). Here we show that 6 out of 16 melanoma tumours analysed acquired EGFR expression after the development of resistance to BRAF or MEK inhibitors. Using a chromatin-regulator-focused short hairpin RNA (shRNA) library, we find that suppression of sex determining region Y-box 10 (SOX10) in melanoma causes activation of TGF-β signalling, thus leading to upregulation of EGFR and platelet-derived growth factor receptor-β (PDGFRB), which confer resistance to BRAF and MEK inhibitors. Expression of EGFR in melanoma or treatment with TGF-β results in a slow-growth phenotype with cells displaying hallmarks of oncogene-induced senescence. However, EGFR expression or exposure to TGF-β becomes beneficial for proliferation in the presence of BRAF or MEK inhibitors. In a heterogeneous population of melanoma cells having varying levels of SOX10 suppression, cells with low SOX10 and consequently high EGFR expression are rapidly enriched in the presence of drug, but this is reversed when the drug treatment is discontinued. We find evidence for SOX10 loss and/or activation of TGF-β signalling in 4 of the 6 EGFR-positive drug-resistant melanoma patient samples. Our findings provide a rationale for why some BRAF or MEK inhibitor-resistant melanoma patients may regain sensitivity to these drugs after a 'drug holiday' and identify patients with EGFR-positive melanoma as a group that may benefit from re-treatment after a drug holiday.

Long-term impact of the 70-gene signature on breast cancer outcome

Several studies have validated the prognostic value of the 70-gene prognosis signature (MammaPrint^R), but long-term outcome prediction of these patients has not been previously reported. The follow-up of the consecutively treated cohort of 295 patients (<53 years) with invasive breast cancer (T1-2N0-1M0; n = 151 N0, n = 144 N1) diagnosed between 1984 and 1995, in which the 70-gene signature was previously validated, was updated. The median follow-up for this series is now extended to 18.5 years. A significant difference is seen in long-term distant metastasis-free survival (DMFS) for the patients with a low- and a high-risk 70-gene signature (DMFS p < 0.0001), as well as separately for node-negative (DMFS p < 0.0001) and node-positive patients (DMFS p = 0.0004). The 25-year hazard ratios (HRs) for all patients for DMFS and OS were 3.1 (95 % CI 2.02–4.86) and 2.9 (95 % CI 1.90–4.28), respectively. The HRs for DMFS and OS were largest in the first 5 years after diagnosis: 9.6 (95 % CI 4.2–22.1) and 11.3 (95 % CI 3.5–36.4), respectively. The 25-year HRs in the subgroup of node-negative patients for DMFS and OS were 4.57 (95 % CI 2.31–9.04) and 4.73 (95 % CI 2.46–9.07), respectively, and for node-positive patients for DMFS and OS were 2.24 (95 % CI 1.25–4.00) and 1.83 (95 % CI 1.07–3.11), respectively. The 70-gene signature remains prognostic at longer follow-up in patients <53 years of age with stage I and II breast cancer. The 70-gene signature’s strongest prognostic power is seen in the first 5 years after diagnosis.

Molecular subtyping of colorectal cancer to identify a mesenchymal tumor type that might benefit from TGF-beta pathway inhibition

456

Background: Currently, most colorectal cancer (CRC) patients receive chemotherapy treatment, even though many patients do not benefit. Therefore, a better understanding of the biology is required to identify those patients who will benefit from chemotherapy and to find a more tailored therapy plan for all other patients. Methods: A molecular subtype classification was developed using full genome expression data of 188 stage I-IV CRC patients and validated in 543 stage II and III patients. Subtypes were correlated to clinical factors, prognosis and treatment benefit (stage III). To determine whether TGF-β signaling is elevated in the tumors, 78 patient biopsies were analyzed for p-SMAD2/3 expression using immunohistochemistry. To analyze the effect of TGF-β activation, we studied the effects of MED12 suppression in SKCO-1 CRC cells under treatment with Cisplatin, Oxaliplatin or 5-FU. Results: We developed a diagnostic test that allows the classification of colorectal cancer tumors into different intrinsic molecular subtypes (A-, B-, C-type). The heterogeneity of these subtypes is largely based on 3 biological hallmarks of the tumor: an epithelial-to-mesenchymal transition, deficiency in mismatch repair genes resulting in a high mutation frequency associated with MSI, and cellular proliferation. Especially the C-type (~15% of CRC tumors) is of clinical interest, as C-type patients have the worst outcome, a mesenchymal phenotype and show no benefit from chemotherapy treatment in our patient set or a public dataset. The C-type subgroup has elevated TGF-β signaling, as shown by TGF-beta and TGF-beta receptor over-expression (TGFB1, p=0.0012; TGFBR1, p=0.0005) and increased phopho-SMAD2/3 staining in the tumor cells (1.9-fold, p=0.0002). In cell line experiments, we show that up-regulation of TGF-β signaling by MED12 knockdown resulted in resistance against chemotherapy by preventing apoptosis. Conclusions: The molecular subtypes differ largely in prognosis and response to chemotherapy. A treatment strategy combining standard drugs with agents suppressing TGF-β signaling might benefit C-type patients.

Performance characteristics of the MammaPrint® breast cancer diagnostic gene signature

BACKGROUND

The analytical performance of multigene signatures depends on many parameters, including precision, repeatability, reproducibility and intratumor heterogeneity. Indicators such as sensitivity, specificity, positive predictive value and negative predictive value are typically used to define the clinical performance of a diagnostic test.

AIM

Here we study these performance characteristics of the MammaPrint^® (Agendia NV, Amsterdam, The Netherlands) 70-gene signature using the US FDA-recommended guidelines, as well as predetermined acceptance criteria.

RESULTS

The clinical and analytical performance characteristics show that MammaPrint is a robust, reproducible, precise test, with a maximum variation of 5% in multiple samplings of the same tissue.

CONCLUSION

MammaPrint is a reliable indicator of distant metastasis in early-stage breast cancer patients of all ages and is well suited for personalized medical care.

Abstract A130: Joint cell-line (CCLE, Sanger) and patient (TCGA) modeling of drug sensitivity reveals novel molecular biomarkers for targeted therapy and conventional chemotherapy

Background: The unresponsiveness to anticancer drugs in patients outlines the need to identify novel and robust biomarkers of response to therapy. The recent release of large molecular cell line datasets labeled for drug sensitivity enables the development of such predictors. Cell line based models poorly reflect the extent of molecular heterogeneity and of tumor microenvironment relationships existing at the patient level. We aimed to overcome these limitations by combining cell line datasets with patient cohorts to uncover novel molecular traits associated with drug response at the patient level.

Methods: For building our model, we used the following publicly available datasets: CCLE (1057 cell lines, 24 drugs), Sanger (790 cell lines, 138 drugs) and TCGA (patients: breast cancer, colorectal cancer, lung squamous carcinoma, lung adenocarcinoma, ovarian cancer, kidney cancer and melanoma). A joint semi-supervised trained Elastic Net approach was used to model drug sensitivity (both conventional chemotherapy and targeted therapy) using gene-expression, copy-number alteration, and somatic mutation data. We introduced a false discovery rate estimation for selecting significant aberrations. As validation, we used pooled RNAi screens (NKI, Netherlands) and also a dataset of 120 patients prospectively enrolled in the ongoing MOSCATO molecular screening program (Gustave Roussy, France) (presented at ASCO 2013).

Results: Drug sensitivity models trained in the entire set of cell lines performed better than models trained using tissue specific cell lines, whatever the drug and the tissue type. Novel molecular traits (mutation, copy number variation) were identified by selecting highly performant models (AUC &gt; .75) across drug and tissue type combinations. We were able to recover all of the recognized landmark biomarkers across drug - tissue combination (e.g. Erlotinib: EGFR mutation in lung adenocarcinoma; Lapatanib: ERBB2 amplification in breast cancer; Vemurafenib: BRAF mutation in melanoma, etc). Moreover, we uncovered novel molecular traits (copy, number mutations) associated (FDR&lt;25%) with drug response (e.g. Erlotinib: CDKN2A deletion in lung adenocarcinoma; Lapatanib: CDH1 deletion in breast cancer; Vemurafenib: FH amplification in melanoma, etc.). The validation of our framework in the pooled RNAi screens of colorectal and lung adenocarcinoma cell lines and in the MOSCATO patient's cohort will be presented.

Conclusions: We provide a comprehensive annotation of mutations and copy-number events in tumors predicted to be associated with sensitivity and resistance in 142 anti-cancer drugs and across 7 tumor types. These molecular traits can be used to refine the results of large scale functional genomic experiments such as RNAi screens and ultimately in the early clinical trial setting.

Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A130.

Citation Format: Charles Ferté, Elias Chaibub Neto, Chong Sun, Cecilia Noecker, Frederic Commo, Olga Nikolova, In Sock Jang, Mehmet Gonen, Benjamin Besse, Fabrice André, Eric Angevin, Ludovic Lacroix, Clement Mazoyer, Antoine Hollebecque, Christophe Massard, Adam Margolin, Roderick Beijersbergen, Stephen Friend, Rene Bernards, Jean-Charles Soria, Justin Guinney. Joint cell-line (CCLE, Sanger) and patient (TCGA) modeling of drug sensitivity reveals novel molecular biomarkers for targeted therapy and conventional chemotherapy. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A130.

Colorectal cancer intrinsic subtypes predict chemotherapy benefit, deficient mismatch repair and epithelial-to-mesenchymal transition

In most colorectal cancer (CRC) patients, outcome cannot be predicted because tumors with similar clinicopathological features can have differences in disease progression and treatment response. Therefore, a better understanding of the CRC biology is required to identify those patients who will benefit from chemotherapy and to find a more tailored therapy plan for other patients. Based on unsupervised classification of whole genome data from 188 stages I–IV CRC patients, a molecular classification was developed that consist of at least three major intrinsic subtypes (A-, B- and C-type). The subtypes were validated in 543 stages II and III patients and were associated with prognosis and benefit from chemotherapy. The heterogeneity of the intrinsic subtypes is largely based on three biological hallmarks of the tumor: epithelial-to-mesenchymal transition, deficiency in mismatch repair genes that result in high mutation frequency associated with microsatellite instability and cellular proliferation. A-type tumors, observed in 22% of the patients, have the best prognosis, have frequent BRAF mutations and a deficient DNA mismatch repair system. C-type patients (16%) have the worst outcome, a mesenchymal gene expression phenotype and show no benefit from adjuvant chemotherapy treatment. Both A-type and B-type tumors have a more proliferative and epithelial phenotype and B-types benefit from adjuvant chemotherapy. B-type tumors (62%) show a low overall mutation frequency consistent with the absence of DNA mismatch repair deficiency. Classification based on molecular subtypes made it possible to expand and improve CRC classification beyond standard molecular and immunohistochemical assessment and might help in the future to guide treatment in CRC patients.

Estrogen receptor splice variants as a potential source of false-positive estrogen receptor status in breast cancer diagnostics

It is well established that only estrogen receptor (ER)-positive tumors benefit from hormonal therapies. We hypothesized that a subgroup of breast cancer patients expresses estrogen receptor α (ERα), but fails to respond to hormonal therapy due to the expression of a non-functional receptor. We analyzed a series of 2,658 ERα-positive HER2-negative breast tumors for ERα and progesterone receptor (PR) status as determined by mRNA expression and for their molecular subtypes (Luminal type vs Basal type, assessed by BluePrint™ molecular subtyping assay). In addition, we assessed the recurrence risk (low vs high) using the 70-gene MammaPrint™ signature. We found that 55 out of 2,658 (2.1 %) tumors that are ERα positive by mRNA analysis also demonstrate a Basal molecular subtype, indicating that they lack expression of estrogen-responsive genes. These ERα-positive Basal-type tumors express significantly lower levels of both ERα and PR mRNA as compared to Luminal-type tumors (P < 0.0001) and almost invariably (94.5 %) have a high-risk MammaPrint™ profile. Twelve of the MammaPrint™ genes are directly ERα responsive, indicating that MammaPrint™ assesses ERα function in breast cancer without considering ERα mRNA levels. We find a relatively high expression of the dominant negative ERα splice variant ERΔ7 in ERα-positive Basal-type tumors as compared to ERα-positive Luminal-type tumors (P < 0.0001). Expression of the dominant negative ERα variant ERΔ7 provides a rationale as to why tumors are of the Basal molecular subtype while staining ERα positive by immunohistochemistry. These tumors may lack a functional response to estrogen and consequently may not respond to hormonal therapy. Our data indicate that such patients are of MammaPrint™ high recurrence risk and might benefit from adjuvant chemotherapy.

Association of colorectal cancer intrinsic subtypes with prognosis, chemotherapy response, deficient mismatch repair, and epithelial to mesenchymal transition (EMT)

3530

Background: Unbiased genome-wide analyses of gene expression patterns have been successfully used for molecular classification of breast cancer into subtypes that have clear relevance for prognosis and treatment. A similar classification is still missing for colorectal cancer (CRC). Methods: Using full genome expression data of 188 stage I-IV CRC patients, an unsupervised clustering revealed three major subtypes (A-, B-, C-type). A molecular subtype classification was developed and validated in 543 stage II and III patients. The subtypes were analyzed for correlation to clinical information, mutations in the kinome, known molecular markers status and chemotherapy response. In addition, subtypes were determined on 173 samples from The Cancer Genome Atlas (TCGA) colon dataset with Agilent genome expression data. Results: C-type patients have the worst outcome, a mesenchymal phenotype, and show no benefit from adjuvant chemotherapy treatment. Patients having A- or B-type tumors have a better clinical outcome, a more proliferative and epithelial phenotype and benefit from adjuvant chemotherapy. A- and C-type groups are enriched for tumors having oncogenic BRAF mutations and a deficient DNA mismatch repair system. B-type tumors showed a low overall kinome mutation frequency (1.6%), while both A- and C-type patients harbor a higher mutation frequency (respectively 4.2 and 6.2%), in agreement with their mismatch repair deficiency. Finally, CRC subtyping was confirmed in the colon TCGA dataset with 26 samples classified as A-type, 110 as B-type and 37 as C-type. In agreement with the different aggressiveness of the subtypes, A-type tumors were less prevalent in stage IV while C-type were less prevalent in stage I CRC. Conclusions: The heterogeneity of the intrinsic subtypes is largely based on three biological hallmarks of the tumor: an epithelial-to-mesenchymal transition, deficiency in mismatch repair genes that result in a high mutation frequency associated with MSI, and cellular proliferation. These subtypes are clinically relevant, as they differ in their underlying biology and might require different treatment strategies.

Next-generation sequencing to find predictors for chemotherapy response in triple-negative breast cancer (TNBC)

1010

Background: In TNBC initial response to chemotherapy is often favorable, but relapse and chemotherapy resistance frequently occur in advanced disease. Hence there is an urgent need for targeted treatments in this breast cancer subtype. Methods: To identify biomarkers of chemotherapy resistance and putative directed treatment targets, we performed next generation sequencing (NGS) of 2,000 genes implicated in oncogenesis. DNA from 31 pre-treatment biopsies and matched normal blood was sequenced. Biopsies were derived from patients scheduled to receive neoadjuvant chemotherapy with doxorubicin/cyclophosphamide. For the analysis, the tumors were divided in responders and non-responders, depending on whether or not a pathological complete remission (pCR) was achieved. Two definitions of pCR were employed: either the complete absence of infiltrating tumor cells in the breast (n=18) or a pCR of both the breast and lymph nodes (n=15). Tumors with partial or no responses were grouped in the non-responder category. Results: As a positive control, we verified that all patients with a known germline BRCA1 mutation (n=8) could be detected by the NGS analysis. In further analyses, we focused only on somatic mutations. Overall, the mutation rate was slightly higher in the non-responders (per tumor, average=12, range=[3-25]) than in the responders (average=8, range=[3-19]) (p=0.17). The analysis of individual genes did not reveal significant differences between responders and non-responders. However, pathway analysis showed that mutations in phosphatidylinositol signaling (e.g., PIK3CA, CALML5) were significantly more frequent in the non-responders, with mutations present in 10/13 non-responders and 2/18 responders (adjusted p=0.013). The chemokine and integrin signaling pathways also revealed a significantly higher mutation rate in the non-responders. Conclusions: Mutations in genes of the phosphatidylinositol pathway occur frequently in TNBCs that do not achieve a pCR on a neoadjuvant chemotherapy regimen consisting of doxorubicin and cyclophosphamide. After validation, alternative chemotherapy regimens and targeted agents should be investigated for these tumors.

Abstract 1218: Development of comprehensive molecular portraits of lobular breast cancer within the RATHER (Rational Therapy for Breast Cancer) Consortium

Background: The RATHER Consortium aims to identify novel kinase targets for therapy in poor-prognosis subtypes of breast cancer for which there are currently no targeted therapies available; namely invasive lobular carcinomas (ILC) which represent 10% of breast tumors.

The RATHER approach differs in two fundamental ways from the ongoing cancer genome re-sequencing efforts. Firstly, the major cancer genome re-sequencing efforts focus on DNA sequence analysis of the whole genome. In RATHER, we identify and validate novel kinases as therapy targets in a more comprehensive way by analyzing the DNA sequence and by applying state-of-the-art proteomics, copy number analysis and gene expression profiling technologies to identify activated/altered kinases. Secondly, RATHER also differs by focusing on the lobular subtype of breast cancer.

Methods: One hundred and fifty ILC samples (fresh frozen) with &gt;5years follow-up were collected from two institutes (Cambridge, NKI). All samples were processed following one standard operating protocol to isolate RNA, DNA and protein of high quality. We used a five-pronged approach to identify and validate novel kinase targets for therapy in ILC breast cancer, namely i) direct re-sequencing of the kinome of 150 ILC breast tumors, ii) determination of abundance and activation status of kinases in these tumors by reverse phase protein lysate array (RPPA) technology, iii) determination of copy number aberration (CNA) in kinase genes by SNP arrays, iv) mRNA quantitation of both genome and kinome using DNA microarrays and v)RNAseq of a subset of ILC tumors.

Results: Data from these independent genome-scale technologies were integrated, yielding a priority list of potential kinase targets for therapy in ILC breast cancer. Deep sequencing of the kinome has revealed somatic mutations characteristic of ILC, which are currently being validated via mass spectrometry-based genotyping technology and their possible effects confirmed with gene expression, protein expression and phosphorylation changes. In addition, a subset of the ILC samples was assessed via RNA sequencing approach to confirm expression of particular mutants. Known gene mutations in ILC such as loss of CDH1 were confirmed; furthermore, a high frequency of PI3K pathway alterations (50%) was observed. Gene expression analysis of such a large set of ILC can reveal subsets of these breast cancers that significantly regulate alternate biological processes. Such biological subsets are currently being validated with clinical and follow-up data.

Conclusion: The RATHER project aims to deliver proof-of-concept for novel therapeutic interventions, together with companion molecular diagnostic assays for patient stratification, for up to 10% of breast cancer patients, where current treatment options are unsatisfactory.

Citation Format: Justine K. Peeters, Ian Majewski, Leanne de Koning, Yue Fan, Finbarr Tarrant, Darran O'Connor, Jeroen Heijmans, Mireille Snel, Tesa Severson, Astrid Bosma, Magali Michaut, Lorenza Mittempergher, Suet-Feung Chin, Thierry Dubois, William Gallagher, Carlos Caldas, Rene Bernards, Iris Simon, RATHER Consortium. Development of comprehensive molecular portraits of lobular breast cancer within the RATHER (Rational Therapy for Breast Cancer) Consortium. [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 1218. doi:10.1158/1538-7445.AM2013-1218

A combined oncogenic pathway signature of BRAF, KRAS and PI3KCA mutation improves colorectal cancer classification and cetuximab treatment prediction

OBJECTIVE

To develop gene expression profiles that characterise KRAS-, BRAF- or PIK3CA-activated- tumours, and to explore whether these profiles might be helpful in predicting the response to the epidermal growth factor receptor (EGFR) pathway inhibitors better than mutation status alone.

DESIGN

Fresh frozen tumour samples from 381 colorectal cancer (CRC) patients were collected and mutations in KRAS, BRAF and PIK3CA were assessed. Using microarray data, three individual oncogenic and a combined model were developed and validated in an independent set of 80 CRC patients, and in a dataset from metastatic CRC patients treated with cetuximab.

RESULTS

175 tumours (45.9%) harboured oncogenic mutations in KRAS (30.2%), BRAF (11.0%) and PIK3CA (11.5%). Activating mutation signatures for KRAS (75 genes), for BRAF (58 genes,) and for PIK3CA (49 genes) were developed. The development of a combined oncogenic pathway signature-classified tumours as 'activated oncogenic', or as 'wildtype-like' with a sensitivity of 90.3% and a specificity of 61.7%. The identified signature revealed other mechanisms that can activate ERK/MAPK pathway in KRAS, BRAF and PIK3CA wildtype patients. The combined signature is associated with response to cetuximab treatment in patients with metastatic CRC (HR 2.51, p<0.0009).

CONCLUSION

A combined oncogenic pathway signature allows the identification of patients with an active EGFR-signalling pathway that could benefit from downstream pathway inhibition.

Identification of synthetic lethal interactions with the BRAF oncogene in colorectal cancer

403

Background: Approximately 8-15% of colorectal (CRC) patients carry an activating mutation in BRAF. This CRC subtype is associated with poor outcome and with resistance, both to chemotherapeutic treatments and to tailored drugs. We recently showed that BRAF (V600E) colon cancers (CCs) have a characteristic gene expression signature (1, 2) which is found also in subsets of KRAS mutant and KRAS-BRAF wild type (WT2) tumors. Tumors having this gene signature, referred as “BRAF-like”, have a similar poor prognosis irrespective of the presence of the BRAF (V600E) mutation. By using a shRNA-based genetic screen in BRAF mutant CC cell lines we aimed to identify genes and pathways necessary for survival and growth of BRAFmutant CC. Such studies may reveal additional targets for therapy and potentially provide new biomarkers for patient stratification Methods: We identified 363 genes that are selectively overexpressed in BRAF mutant tumors as compared to WT2 type tumors, based on gene expression profiles of the PETACC3 (1) and Agendia (2) datasets. The TRC human genome-wide shRNA collection (TRC-Hs1.0) was used to generate a 1815 hairpins sub-library targeting those identified genes (BRAF library). BRAF(V600E) CC cell lines were infected with the BRAF library and screened for shRNAs that cause lethality. LIM1215 CC cell line (WT2) was used as a control. Cells stably expressing the shRNA library were cultured for 13 days, after which shRNAs were recovered by PCR. Deep sequencing was applied to determine the specific depletion of shRNA in BRAF(V600E) cells as compared to LIM1215 cells Results: Candidate genes were identified by using following filtering criteria: depletion in BRAF(V600E) cells by at least 50% and depletion in BRAF(V600E) cells 1, 5-fold higher than in control cells with the corresponding p-value to be ≤ 0.1. A total of 34 genes met our criteria of which 6 genes were presented with more than one hairpin and were concordant across the cell lines selected for validation. Conclusions: We identified candidate synthetic lethal genes in BRAF mutant CC cell lines. Functional analysis is ongoing. Data will be presented. References 1. J Clin Oncol 2012 Apr 20;30(12):1288-9 2. Gut (2012). doi:10.1136/gutjnl-2012-302423

Association of colorectal cancer intrinsic subtypes with prognosis, chemotherapy response, deficient mismatch repair, and epithelial to mesenchymal transition (EMT)

333

Background: Unbiased genome-wide analyses of gene expression patterns have been successfully used for the molecular classification of breast cancer into subtypes that have clear relevance for prognosis and development of treatment plans. For colorectal cancer (CRC), however, a molecular classification is still missing. Methods: Using gene expression data of 188 stage I-IV colorectal cancer (CRC) patients, a molecular subtype classification was developed. The classifier was validated in 543 stage II and III patients and the subtypes were analyzed for correlation to clinical information, mutations in the kinome, known molecular marker status and chemotherapy response. Results: CRC is a heterogeneous disease that consists of at least three major intrinsic subtypes (A-, B-, C-type). The heterogeneity of the intrinsic subtypes is largely based on three biological hallmarks of the tumor: an epithelial-to-mesenchymal transition, deficiency in mismatch repair genes that result in a high mutation frequency associated with MSI, and cellular proliferation. C-type patients have the worst outcome, a mesenchymal gene expression phenotype, and show no benefit from adjuvant chemotherapy treatment. Patients having A-type or B-type tumors have a better clinical outcome, a more proliferative and epithelial phenotype, and benefit from adjuvant chemotherapy. B-type tumors showed a low overall kinome mutation frequency (1.6%), while both A-type and C-type patients harbor a higher mutation frequency (respectively 4.2 and 6.2%), in agreement with their mismatch repair deficiency. Conclusions: We have developed a diagnostic single sample predictor that allows the classification of CRC tumors of different intrinsic molecular subtypes. These subtypes are potentially clinically relevant, as they differ in their underlying biology and clinical outcome and consequently require different treatment strategies. [Table: see text]