A chemogenomic approach to identify personalized therapy for patients with relapse or refractory acute myeloid leukemia: results of a prospective feasibility study

Targeted next-generation sequencing (tNGS) and ex vivo drug sensitivity/resistance profiling (DSRP) have laid foundations defining the functional genomic landscape of acute myeloid leukemia (AML) and premises of personalized medicine to guide treatment options for patients with aggressive and/or chemorefractory hematological malignancies. Here, we have assessed the feasibility of a tailored treatment strategy (TTS) guided by systematic parallel ex vivo DSRP and tNGS for patients with relapsed/refractory AML (number NCT02619071). A TTS issued by an institutional personalized committee could be achieved for 47/55 included patients (85%), 5 based on tNGS only, 6 on DSRP only, while 36 could be proposed on the basis of both, yielding more options and a better rationale. The TSS was available in <21 days for 28 patients (58.3%). On average, 3 to 4 potentially active drugs were selected per patient with only five patient samples being resistant to the entire drug panel. Seventeen patients received a TTS-guided treatment, resulting in four complete remissions, one partial remission, and five decreased peripheral blast counts. Our results show that chemogenomic combining tNGS with DSRP to determine a TTS is a promising approach to propose patient-specific treatment options within 21 days.

Publisher Correction: SLX4 interacts with RTEL1 to prevent transcription-mediated DNA replication perturbations

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Adolescents and young adults with classical Hodgkin lymphoma in northern Tunisia: insights from an adult single-institutional study

PURPOSE

The aim of this study was to extensively describe the epidemiological, clinical and therapeutic outcomes of adolescents and young adults (AYA) population with classical Hodgkin Lymphoma (cHL). Then, a comparison between AYAs and adults and between the subgroups of AYAs treated with the same adult protocol was accomplished to further inform on optimal therapy approach of choice for adolescent patients.

MATERIAL AND METHODS

In this mono-centric, retrospective study, we reviewed the medical records. We analyzed 112 consecutive North Tunisian patients, including 66 AYAs (15 to 39 years) and 46 adults (≥40years) affected by cHL treated from 2000 to 2015 at Salah Azaiez Institute. Then, we performed a comparative analysis between AYA and 46 adult patients and a subgroup analysis between adolescents and young adults. All patients were treated according to the national protocol for HL, edited by the Tunisian Society of Hematology. The treatment included chemotherapy and involved-field radiotherapy (RT) at a dose of 20 or 30 Grays (Gy) for responders and 36Gy for non-responders.

RESULTS

AYA patients presented with adverse features with nodular sclerosis subtype (p=3.88×10^-02) and mediastinal mass involvement (p=9.40×10^-04). At a median follow-up of 51 and 32 months for AYAs and adults, respectively, no statistical difference in terms of 3 and 5-years overall survival (OS) and event-free survival (EFS) was shown. Using the Kaplan-Meier method, in AYAs, the ABVD regimen has an impact on 3-years EFS (p=4.63×10^-02). The 36Gy RT was associated with the best 3-years EFS (p=9.24×10^-03). Besides, AYA patients with advanced-stage had the worst 3-years OS (76%) (p=2.41×10^-02). Although the adolescents and young adults shared similar clinical presentation, we noted that the adolescent group had the worst 3-years EFS (48%), but the best 3-years OS (91%). We identified 15% of primary refractory patients and a rate of toxicity of 5.3% in AYA.

CONCLUSION

The treatment approach used is well tolerated by adult patients. However, the AYA patients and particularly adolescent subgroup had more advanced disease at diagnosis and should be treated more intensively in dedicated units. RT dose<36Gy and ABVD chemotherapy were associated with lower EFS in this population.

Abstract P5-06-02: Ex vivo CSC assays for personalized testing of drug susceptibility in advanced breast cancer

In the developing area of personalized medicine, targeted therapies are mainly based on genomic characterization of each tumor, and is currently proposed as promising strategies for advanced breast cancer (ABC). Despite the promises of advanced genome sequencing, many patients still fail therapy, resulting in disease progression, recurrence, and metastases. Cancer stem cells (CSCs) concept illustrates the non-genetic intrinsic resistance, recapitulates tumor heterogeneity that creates hierarchically organized tumor tissues where a subpopulation of self-renewing cancer stem cells (CSCs) sustains the long- term clonal maintenance of the neoplasm. Evidences indicate that CSCs survive many commonly employed cancer therapeutics. Patient-derived tumor xenograft (PDXs) models recapitulate tumor complexity and heterogeneity at cellular and molecular level.

We aimed to specifically address the therapeutic sensitivity in ABC, by using an ex vivo assay based on PDX prospective collection, fully characterized for genomic alterations.

In this work, we aim at defining for each tumor the best therapy to target breast cancer intratumor heterogeneity, the CSC component. For that, we defined a panel of 44 FDA-approved compounds used for cancer treatment, including breast and other types of cancer, cancer stem cell drugs, chemo or targeted therapies. For each drug, we screened the differential sensitivity of the bulk tumor cells and the CSC components for 12 PDX models using an ex vivo screening approach on short term culture. To assess intra tumor heterogeneity, we set up an original dual strategy: for the bulk cells, an ex vivo assay based on IC50, and for breast CSC component a miniaturized Aldefluor assay. First, we demonstrate that bulk cells and CSCs sensitivity may be dissociated for the same drug in the same PDX models. Then, we observed that whereas bulk cell sensitivity may be correlated to tumor genomic abnormalities, CSC drug sensitivity seems not to follow the rule.CSC are selectively sensitive to specific compounds. We are exploring the pathways that sustain this selective sensitivity in the CSCs components. We are currently identifying targets using mass spectrometry in CSCs and bulk cells.Then, we validated the hits predicted from ex vivo screening assays by in vivo treatment of using PDX models for the selected drugs, and in a patient with ABC.

In that work, we demonstrated that CSCs display different sensitivity profiles than bulk cells to the same agents, irrespective to their genomic background and are identifying the CSC specific targets. Here, we propose a new model of precision medicine based on ex vivo CSC assays for personalized testing of drug susceptibility in advanced breast cancer.

Citation Format: Charafe-Jauffret E, Wicinski J, Cabaud O, Lopez M, Audebert S, Adelaide J, Chaffanet M, Guille A, Goncalves A, Bertucci F, Birnbaum D, Ginestier C. Ex vivo CSC assays for personalized testing of drug susceptibility in advanced breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P5-06-02.

Nectin-4: a new prognostic biomarker for efficient therapeutic targeting of primary and metastatic triple-negative breast cancer

Background

Triple-negative breast cancers (TNBCs) are associated with a poor prognosis. In contrast to other molecular subtypes, they have no identified specific target and chemotherapy remains the only available systemic treatment. The adhesion molecule nectin-4 represents a new potential therapeutic target in different cancer models. Here, we have tested the prognostic value of nectin-4 expression and assessed the therapeutic efficiency of an anti-nectin 4 antibody drug conjugate (ADC) on localised and metastatic TNBC in vitro and in vivo.

Materials and methods

We analysed nectin-4/PVRL4 mRNA expression in 5673 invasive breast cancers and searched for correlations with clinicopathological features including metastasis-free survival (MFS). Immunohistochemistry was carried out in 61 TNBCs and in samples of primary TNBC Patient-Derived Xenografts (PDXs). An anti-nectin-4 antibody eligible for ADC was produced and tested in vitro and in vivo in localised and metastatic TNBC PDXs.

Results

High nectin-4/PVRL4 mRNA expression was associated with poor-prognosis features including the TN and basal subtypes. High PVRL4 mRNA expression showed independent negative prognostic value for MFS in multivariate analysis in TNBCs. Nectin-4 protein expression was not detected in adult healthy tissues including mammary tissue. Membranous protein expression was found in 62% of TNBCs, with strong correlation with mRNA expression. We developed an ADC (N41mab-vcMMAE) comprising a human anti-nectin-4 monoclonal antibody conjugated to monomethyl auristatin-E (MMAE). In vitro, this ADC bound to nectin-4 with high affinity and specificity and induced its internalisation as well as dose-dependent cytotoxicity on nectin-4-expressing breast cancer cell lines. In vivo, this ADC induced rapid, complete and durable responses on nectin-4-positive xenograft TNBC samples including primary tumours, metastatic lesions, and local relapses; efficiency was dependent on both the dose and the nectin-4 tumour expression level.

Conclusion

Nectin-4 is both a new promising prognostic biomarker and specific therapeutic target for ADC in the very limited armamentarium against TNBC.

Abstract P4-13-23: Next-generation sequencing (NGS), array comparative genomic hybridization (aCGH) and patient-derived tumor xenograft (PDX) for precision medicine in advanced breast cancer: A single-center prospective study

Background

Genomic-based approaches in advanced breast cancer (ABC) were recently demonstrated as feasible in the clinical practice, but only a limited number of patients were actually treated with targeted therapies matching genomic alterations, with low antitumor activity. We conducted a pilot study to evaluate whether precision medicine using NGS and aCGH could be implemented prospectively at a single center in ABC patients. In addition, we examined whether PDX could be derived from ABC and thus could help inform therapeutic decision.

Methods

ABC patients accessible to tumor biopsy were prospectively enrolled at the Institut Paoli-Calmettes in the BC-BIO study (ClinicalTrials.gov, NCT01521676). Tumor tissue from locally recurrent or metastatic disease was immediately frozen after dedicated biopsy. Genomic profiling included high-resolution 4x180K aCGH (Agilent Technologies, Massy, France) and DNA sequencing, using a library of 365 cancer candidate genes (HaloPlex target enrichment kit, Agilent technologies, Santa Clara, CA, USA) and MiSeq analyzer (Illumina, San Diego, CA, USA) with 2x150-bp, paired-end at about 300x coverage. In a subset of patients, fresh tumor was implanted orthotopically in humanized cleared fat pads of NSG mice for establishing xenotransplants.

Results

A total of 34 ABC patients were included, with the following characteristics: median age 54 years (35-77); molecular subtypes: 11 triple-negative (32%), 12 luminal non-HER2 (35%), 4 luminal HER2 (12%), 3 HER2 non-luminal (9%), and 4 unknown (12%); 33 with previous chemotherapy (97%); 22 with previous endocrine treatment (35%); 7 with previous anti-HER2 (21%). Tumor biopsies were obtained from liver (15), skin (6), peritoneum (4), breast (3), node (3), lung (1), pleura (1), and ascitis (1), with a median tumor cellularity of 70% (range 10-90%). aCGH and NGS were available from 34 and 33 patients, respectively. An actionable target was found in 28 patients (82%), corresponding to 66 targets, including 37 mutations (8 in PIK3CA, 7 TP53, 4 ESR1, 2 AKT1, 2 BRCA2, 2 HER2), 22 amplifications (7 for CCND1, 2 CCNE1, 2 FGFR1, 2 IGF1R) and 7 homozygous deletions (3 for PTEN, 2 CDKN2A/B,1 BRCA2, 1 STK11). A targeted therapeutic proposal was possible, either in a clinical trial (N=18, 52%) or using already registered drugs (N=17, 50%). Ten patients actually received a targeted treatment, 1 of them experienced objective response and 1 showed stable disease for more than 6 months. Of 26 patients subjected to mouse implantation, 10 had successful xenografting (6 triple-negative, 2 HER2, 1 luminal non-HER2, 1 subtype non-attributed), with a median time to reach 10 mm of 148 days. These PDX will be used as models to understand the patient's therapeutic response.

Conclusion

Precision medicine using high-throughput DNA sequencing and aCGH can be implemented at a single center in the context of clinical practice and may allow direct therapeutic proposal in 1/3 of patients, but antitumor activity was minimal. PDX may be obtained in a significant fraction of patients, especially in triple-negative and HER2 subtypes, and could phenotypically complement genomic data.

Citation Format: Gonçalves A, Bertucci F, Chaffanet M, Guille A, Garnier S, Adelaide J, Carbuccia N, Brunelle S, Piana G, Cabaud O, Thomassin-Piana J, Paciencia-Gros M, Chereau-Ewald E, Lambaudie E, Sabatier R, Tarpin C, Provansal M, Jalaguier-Coudray A, Extra J-M, Sarran A, Pakradouni J, Viens P, Lopez M, Ginestier C, Charafe-Jauffret E, Birnbaum D. Next-generation sequencing (NGS), array comparative genomic hybridization (aCGH) and patient-derived tumor xenograft (PDX) for precision medicine in advanced breast cancer: A single-center prospective study. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P4-13-23.

Analysis of Hospital Infection Surveillance Data

How often infection rates should be calculated and how large a change is required for “significance” are pertinent questions in nosocomial infection surveillance programs. A method is presented which establishes outbreak threshold infection frequencies. Comparison is direct and immediate: computation of rates or use of electronic data processing is not required. We have validated this method, using computer systems, by comparing the distributions of mean weekly incidence and prevalence statistics for each ward by nosocomial infection site in an acute care general hospital against both our theoretical outbreak threshold limits and the distribution of proven infection outbreaks. Sensitive and specific distinction between random variation or sporadic cross-infection and true persisting outbreaks requiring intervention is obtained. This approach provides a simple and timely alternative to intuitive after-the-fact interpretation of infection patterns which is applicable to infection surveillance and cost-effective infection control in hospitals of all sizes.

Gene expression profiles of inflammatory breast cancer: correlation with response to neoadjuvant chemotherapy and metastasis-free survival

BACKGROUND

Inflammatory breast cancer (IBC) is an aggressive disease. To date, no molecular feature reliably predicts either the response to chemotherapy (CT) or the survival. Using DNA microarrays, we searched for multigene predictors.

PATIENTS AND METHODS

The World IBC Consortium generated whole-genome expression profiles of 137 IBC and 252 non-IBC (nIBC) samples. We searched for transcriptional profiles associated with pathological complete response (pCR) to neoadjuvant anthracycline-based CT and distant metastasis-free survival (DMFS) in respective subsets of 87 and 106 informative IBC samples. Correlations were investigated with predictive and prognostic gene expression signatures published in nIBC (nIBC-GES). Supervised analyses tested genes and activation signatures of 19 biological pathways and 234 transcription factors.

RESULTS

Three of five tested prognostic nIBC-GES and the two tested predictive nIBC-GES discriminated between IBC with and without pCR, as well as two interferon activation signatures. We identified a 107-gene signature enriched for immunity-related genes that distinguished between responders and nonresponders in IBC. Its robustness was demonstrated by external validation in three independent sets including two IBC sets and one nIBC set, with independent significant predictive value in IBC and nIBC validation sets in multivariate analysis. We found no robust signature associated with DMFS in patients with IBC, and neither of the tested prognostic GES, nor the molecular subtypes were informative, whereas they were in our nIBC series (220 stage I-III informative samples).

CONCLUSION

Despite the relatively small sample size, we show that response to neoadjuvant CT in IBC is, as in nIBC, associated with immunity-related processes, suggesting that similar mechanisms responsible for pCR exist. Analysis of a larger IBC series is warranted regarding the correlation of gene expression profiles and DMFS.

KIT-D816V oncogenic activity is controlled by the juxtamembrane docking site Y568-Y570

Mutation of KIT receptor tyrosine kinase at residue D816 results in ligand-independent constitutive kinase activity. This mutation occurs in most patients with mastocytosis, a myeloproliferative neoplasm, and is detected at lower frequencies in acute myeloid leukemia and in germ cell tumors. Other KIT mutations occur in gastrointestinal stromal tumors (GIST) and mucosal melanoma. KIT is considered as a bona fide therapeutic target as c-kit mutations are driving oncogenes in these pathologies. However, several evidences suggest that KIT-D816V mutant is not as aggressive as other KIT mutants. Here, we show that an intracellular docking site in the juxtamembrane region of KIT maintains a negative regulation on KIT-D816V transforming potential. Sixteen signaling proteins were shown to interact with this motif. We further demonstrate that mutation of this site results in signaling modifications, altered gene expression profile and increased transforming activity of KIT-D816V mutant. This result was unexpected as mutations of the homologous sites on wild-type (WT) KIT, or on the related oncogenic FLT3-ITD receptor, impair their function. Our results support the hypothesis that, KIT-D816V mutation is a mild oncogenic event that is sufficient to confer partial transforming properties, but requires additional mutations to acquire its full transforming potential.

Are histopathological findings of diagnostic value in native valve endocarditis?

BACKGROUND

Optimal management of infective endocarditis (IE) depends on the early detection of IE-causing pathogens and on appropriate antimicrobial and surgical therapy. The current guidelines of the European Society of Cardiology (ESC) recommend histopathological examination as the gold standard for diagnosing IE Habib et al. (Eur Heart J 30:2369-2413, 2005). We hypothesize that histopathological findings do not provide additional information relevant to clinical decision-making.

METHODS

We retrospectively reviewed a cohort of patients who had undergone surgery for native valve endocarditis (NVE) at the University Hospital Regensburg between September 1994 and February 2005. All episodes of intraoperatively confirmed endocarditis during this period were included in the study. Data were retrieved from surgical records, microbiological and histopathological reports, and medical files of the treating as well as admitting hospital. Pathogens were correlated with the site of manifestation of the affected heart valve and with clinical and histopathological findings.

RESULTS

A total of 163 episodes of NVE were recorded and entered into our study for analysis. The valves affected were the aortic valve (45 %), the mitral valve (28 %), the aortic and mitral valve (22 %), and other valves (5 %). IE-causing pathogens were Staphylococcus aureus (22 %), viridans streptococci (18 %), enterococci (10 %), streptococci other than Streptococcus viridans (9 %), coagulase-negative staphylococci (5 %), miscellaneous pathogens (4 %), and culture-negative endocarditis (33 %). Infection with S. aureus was associated with high rates of sepsis, septic foci, and embolic events, while patients with enterococcal IE showed the highest rate of abscesses. Mortality rate in all subgroups was low without significant differences. However, histopathological findings correlated poorly with the pathogen involved and showed only few significant associations that were without clinical relevance.

CONCLUSIONS

The clinical presentation of IE depends on the pathogen involved. Among the episodes of NVE examined, the histopathological examination of resected heart valves did not show any pathogen-specific morphological patterns and therefore did not provide any additional information of clinical value. Based on our findings, we recommend complementary cultures of the resected materials (valve tissue, thrombotic material, pacer wire) and implementation of molecular diagnostic methods (e.g., broad-range PCR amplification techniques) instead of histopathological analyses of resected valve tissue.

Genomic grade adds prognostic value in invasive lobular carcinoma

BACKGROUND

The prognostic value of histologic grade (HG) in invasive lobular carcinoma (ILC) remains uncertain, and most ILC tumors are graded as HG2. Genomic grade (GG) is a 97-gene signature that improves the prognostic value of HG. This study evaluates whether GG may overcome the limitations of HG in ILC.

METHODS

Gene expression data were generated from frozen tumor samples, and GG calculated according to the expression of 97 genes. The prognostic value of GG was assessed in a stratified Cox regression model for invasive disease-free survival (IDFS) and overall survival (OS).

RESULTS

A total of 166 patients were classified by GG. HG classified 33 (20%) tumors as HG1, 120 (73%) as HG2 and 12 (7%) as HG3. GG classified 106 (64%) tumors as GG low (GG1), 29 (17%) as GG high (GG3) and 31 (19%) as equivocal (cases not classified as GG1 or GG3). The median follow-up time was 6.5 years. In multivariate analyses, GG was associated with IDFS [HR(GG3 vs GG1) 5.6 (2.1-15.3); P < 0.001] and OS [HR(GG3 vs GG1) 7.2, 95% CI (1.6-32.2); P = 0.01].

CONCLUSIONS

GG outperformed HG in ILC and added prognostic value to classic clinicopathologic variables, including nodal status.

Abstract P5-03-01: Cancer stem cells predict engraftment and poor prognosis of primary breast tumors

Breast cancer is a major health problem and heterogeneity of the disease has been considered as a strong limitation to find the best therapies to cure cancer, overcome recurrences and metastases. The establishment of models that reflect tumor biology and metastatic progression is critical to develop successful new therapeutic strategies. In the breast, orthotopic xenografts currently appear as the best models to study tumor growth, metastasis and develop tools for prognosis prediction. Furthermore, mouse transplant assays have been used to assess cancer stem cell (CSC) activity and demonstrate that leukemia and many solid tumors are organized along a hierarchical model.

Despite the promise of the CSC model sustained by mouse transplantation assays, the clinical relevance of xenografts studies to identify determinants of stemness able to influence clinical outcome remains challenging. In breast cancer, transcriptional programs from functionally validated CSC populations remain to be deciphered.

Here, we report the establishment of a bank of primary breast tumor-derived xenografts (xenobank). We showed that the xenografts retain the main features of primary tumors, that engraftment is correlated with the presence of CSC in tumors, and that engraftment in the mouse is able to predict prognosis in patients. This suggests that CSCs may govern breast cancer prognosis. We established the gene expression profiles of functionally validated ALDEFLUOR-positive CSC populations (breast CSC-GES) and demonstrated their clinical relevance. Among 2609 patients with breast cancers, we validated that he expression of the breast CSC-GES is correlated with poor outcome and metastasis in uni-and multivariate analysis (5-year MFS was 70% CI95 [67–74] in the breast CSC-positive class and 80% (CI95 [77–83]) in the breast CSC-negative class (p = 5.5E−04 with log-rank test). Furthermore, we identified a core of 19 genes commonly expressed in breast CSC, murine embryonic, neural and hematopoietic stem cells programs and demonstrated for each gene its ability to modulate breast CSC population, being implicated in self-renewing or differentiation programs. We found that the core of genes in common between four stem cell gene expression studies (CE-BCSC-3SC) displayed an adverse prognostic impact for patients with breast cancer. The core contained genes implicated in oxidative phosphorylation, detoxification, lipid metabolism, and genomic stability, and these shared determinants of stemness influenced clinical outcome.

Thus, we show ed that CSCs from orthotopically engrafted primary breast tumors have clinical and biological relevance. This functionally validated CSC population is highly correlated with survival and express genes governing main stem cell functions, substantiating a major prediction of the CSC model and opening further promises for new CSC therapies using valid preclinical models.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-03-01.

CXCR7 receptors facilitate the progression of colon carcinoma within lung not within liver

Background:

Liver and lung metastases are the predominant cause of colorectal cancer (CRC)-related mortality. Chemokine-receptor pairs have a critical role in determining the metastatic progression of tumours. Our hypothesis was that disruption of CXCR7/CXCR7 ligands axis could lead to a decrease in CRC metastases.

Methods:

Primary tumours and metastatic tissues from patients with CRC were tested for the expression of CXCR7 and its ligands. Relevance of CXCR7/CXCR7 ligands for CRC metastasis was then investigated in mice using small pharmacological CXCR7 antagonists and CRC cell lines of human and murine origins, which – injected into mice – enable the development of lung and liver metastases.

Results:

Following injection of CRC cells, mice treated daily with CXCR7 antagonists exhibited a significant reduction in lung metastases. However, CXCR7 antagonists failed to reduce the extent of liver metastasis. Moreover, there were subtle differences in the expression of CXCR7 and its ligands between lung and liver metastases.

Conclusion:

Our study suggests that the activation of CXCR7 on tumour blood vessels by its ligands may facilitate the progression of CRC within lung but not within liver. Moreover, we provide evidence that targeting the CXCR7 axis may be beneficial to limit metastasis from colon cancer within the lungs.

Coamplification in human breast-tumors and physical linkage at chromosomal band 12p13, of ccnd2 and fgf6 genes

The CCND2 and FGF6 genes are both located at band p13 of human chromosome 12. The two genes are coamplified in some cases (3.6%) of breast tumors. They are physically linked within a 500 kb genomic segment. Thus, both 11q13 and 12p13 chromosomal regions host related linked cyclin and FGF genes, and can be amplified in human tumors. This strengthens the hypothesis of genome duplication involving chromosomes 11 and 12.

Oncogenesis and genome duplication maps (commentary)

Examples of activation of putative cancer genes belonging to multigene families are given. Their structure and chromosomal location helps in identifying genome duplications. Reciprocally, genome duplication maps may help in predicting the location and behaviour of putative cancer genes.

Allelic loss at chromosome 8p in human breast-cancer

Loss of heterozygosity (LOH) at loci from the short arm of chromosome 8 has been shown to occur in several types of carcinomas. The consensus deletion region has been recently mapped at 8p21-p22, distal to D8S283 and NEFL loci, and proximal to LPL and D8S265 loci. We report LOH at D8S133, a marker located in this consensus region, in a panel of breast tumor samples. LOH at this locus was observed in about 20% of the tumors.

Basal breast cancer: a complex and deadly molecular subtype

During the last decade, gene expression profiling of breast cancer has revealed the existence of five molecular subtypes and allowed the establishment of a new classification. The basal subtype, which represents 15-25% of cases, is characterized by an expression profile similar to that of myoepithelial normal mammary cells. Basal tumors are frequently assimilated to triple-negative (TN) breast cancers. They display epidemiological and clinico-pathological features distinct from other subtypes. Their pattern of relapse is characterized by frequent and early relapses and visceral locations. Despite a relative sensitivity to chemotherapy, the prognosis is poor. Recent characterization of their molecular features, such as the dysfunction of the BRCA1 pathway or the frequent expression of EGFR, provides opportunities for optimizing the systemic treatment. Several clinical trials dedicated to basal or TN tumors are testing cytotoxic agents and/or molecularly targeted therapies. This review summarizes the current state of knowledge of this aggressive and hard-to-treat subtype of breast cancer.

P5-01-01: Identification, Validation and Assessment of Transcriptional Relevance of a PDGFR-Activation Signature in (Inflammatory) Breast Cancer

Introduction. Breast cancer can be divided into several subgroups characterized by unique patterns of pathway activation. Platelet-derived growth factor receptor (PDGFR) signalling has not yet been included in this classification scheme, although it has been reported to be a potential target for therapy. In this study, we have constructed a PDGFR-activation signature and investigated its relevance in breast cancer.

Materials and Methods. Sixteen PDGFR-modulated genes were identified by intersecting two published PDGFR-modulated gene lists. The resulting gene signature was applied onto a publicly available gene expression data set of GIST (GSE17743) using principle component analysis. The segregation of PDGFR- and KIT-mutated GIST samples was investigated using permutation analysis and classification sensitivity and specificity were assessed. Using the regression coefficients from the first principal component, a PDGFR-activation score was constructed and applied onto a second data set in order to validate the score (GSE1923). Finally, the score was applied onto a gene expression data set of 389 breast cancer ***samples, including 137 samples from patients with IBC.

Results. Sixteen PDGFR-modulated genes (NR4A1, EGR3, JUNB, IER3, TIEG, JUN, BCL3, MYC, NR4A3, PLAU, MCL1, DUSP1, DUSP5, DUSP6, SGK, GADD45A) were able to discriminate PDGFR-mutated GIST samples from KIT-mutated GIST samples with a sensitivity of 75% and a specificity of 85%. Application of the PDGFR-activation score onto a data set of control and PDGF-treated glioblastoma cells showed a significant increase in the PDGFR-activation score in the treated condition (P=0.0302). Application of the PDGFR-signature onto our series of IBC and nIBC samples demonstrated a significant and molecular subtype-independent increase in PDGFR-activation in IBC (P=0.0015; FDR=3%). In addition, in our series of nIBC samples only, PDGFR-activation was associated with decreased DMFS and RFS (P=0.0038 and P=0.0137 respectively). In fact, PDGFR-activation was an independent prognosticator in a multivariate model incorporating the molecular subtypes.

Discussion. We identified a gene signature composed of 16 genes able to predict PDGFR-activation in tissue samples by gene expression analysis. PDGFR-activation is significantly increased in samples from patients with IBC, an aggressive form of locally advanced breast cancer. In addition, in nIBC, PDGFR-activation is associated with DMFS and RFS, independently of the molecular subtypes suggesting that PDGFR-activation might add another level of clinically relevant heterogeneity in breast cancer.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-01-01.

P4-09-12: Clinical Validation of Immunohistochemical Signature Predictive of Patients' 8 Year Outcome in Node Negative Breast Carcinomas

Background

Immunohistochemical (IHC) signatures predictive of prognosis can be a cost effective alternative to genomic signature suitable in clinical practice for patients’ management in early stages of breast carcinoma (BrCa) in clinical. IHC procedures are very easy to handle in pathology laboratories, since require very few amount of formalin fixed tissue samples since assessed on consecutive 4 μm thick sections of paraffin embedded tumors fragments validated by pathologists’ expertise. We recently reported (C Charpin et al Int J Cancer 2009) that standardized IHC assays using high throughput densitometry on digitized microscopic images allow to predict an adverse outcome of patients with BrCa, provided that identified the IHC predictive signatures can be validated in other patients’ cohorts. In this regard we have repeated our previously reported IHC procedures on 418 series of BrCa (Marseille University) on a second series including 303 BrCa diagnosed at Yale University.

Material and Methods: TMA(s) of two cohorts of patients with BrCa (418/Marseille University and 303 Yale University) were respectively investigated for immunohistochemical expression of 15 markers, including (i) those of the best signature predictive of patients outcome in N- patients in our previous studies and (ii) new markers reported in the literature as EMT cells and Stem cells. Monoclonal antibodies against HIFa, PI3, pAKT, pmTOR, moesin, P21, 4EBP-1, P27, Ker 5.6, pMAPKAPK-2, SHARP2, Claudin, ALDH, AF6, CD24, IHC was screened using standardized (Ventana Benchmark XT) on consecutive 4 μm thick sections of TMAs from both series, Quantitative measurements of immunoprecipitates were automatically assessed with TRIBVN device, and correlated to 8 year patients’ outcome. Logistic regression determined the best combination of markers to predict patients’ prognosis and results of the two cohorts were compared.

Results: In node negative BrCa the best predictive combination of markers signature was HIF-1a, PI3K, Claudin, AF6, pAKT independently of ER PR and HER-2 status. When results of logistic regression with these markers was compared very similar results were obtained with 92.34 % (333/418) of well classified patients in the first set (Marseille University series) and 89.8 % (158/176) in the second set from Yale University.

Discussion: The results suggest that the new optimal IHC signature predictive of prognosis identified in BrCa including HIF-1a, PI3K, Claudin, AF6 and pAKT is validated by this study with very similar results in two different cohorts of patients.

This validation also suggests (1) that IHC signatures investigated in individual patients can be considered as suitable in clinical practice offering a cost effective convenient mean to predict patients’ outcome at diagnostic time (2) and to select those patients with poor prognosis, particularly in N- BrCa subset, requiring more aggressive adjuvant therapies, and to avoid useless expensive therapies and their side effects in N- patients with favourable prognosis.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-09-12.

A refined molecular taxonomy of breast cancer

The current histoclinical breast cancer classification is simple but imprecise. Several molecular classifications of breast cancers based on expression profiling have been proposed as alternatives. However, their reliability and clinical utility have been repeatedly questioned, notably because most of them were derived from relatively small initial patient populations. We analyzed the transcriptomes of 537 breast tumors using three unsupervised classification methods. A core subset of 355 tumors was assigned to six clusters by all three methods. These six subgroups overlapped with previously defined molecular classes of breast cancer, but also showed important differences, notably the absence of an ERBB2 subgroup and the division of the large luminal ER+ group into four subgroups, two of them being highly proliferative. Of the six subgroups, four were ER+/PR+/AR+, one was ER−/PR−/AR+ and one was triple negative (AR−/ER−/PR−). ERBB2-amplified tumors were split between the ER−/PR−/AR+ subgroup and the highly proliferative ER+ LumC subgroup. Importantly, each of these six molecular subgroups showed specific copy-number alterations. Gene expression changes were correlated to specific signaling pathways. Each of these six subgroups showed very significant differences in tumor grade, metastatic sites, relapse-free survival or response to chemotherapy. All these findings were validated on large external datasets including more than 3000 tumors. Our data thus indicate that these six molecular subgroups represent well-defined clinico-biological entities of breast cancer. Their identification should facilitate the detection of novel prognostic factors or therapeutical targets in breast cancer.

Abstract P4-09-01: Poly (ADP-ribose) Polymerase-1 mRNA Expression in Human Breast Cancer: Correlations with Subtypes and Prognostic Impact

Introduction: Although poly(ADP-ribose) polymerase-1 (PARP1) inhibition is a recent promising therapy in breast cancer, PARP1 expression in this disease is not known.

Methods. Using DNA microarray and array-based comparative genomic hybridization (arrayCGH), we examined PARP1 mRNA expression and copy number alterations in 326 invasive breast cancer samples and normal breast samples. A meta-analysis was performed on a large public retrospective gene expression dataset (n= 2,485) to analyze correlation between PARP1 mRNA expression and molecular subtypes and clinico-pathological parameters.

Results. PARP1 was overexpressed in 58% of cancers, and its expression was heterogeneous between tumors. ArrayCGH data revealed an association between mRNA overexpression and gain/amplification at the PARP1 locus (p&lt;1.0E-8). Meta-analysis showed that PARP1 expression was higher in basal breast cancers (p&lt;1.0E-72), but overexpression was also found in other subtypes. PARP1 expression correlated with high grade, medullary histological type, tumor size and worse metastasis-free survival (HR=1.12 [1.04-1.22]; p=0.004). In multivariate analysis, PARP1 expression had an independent prognostic value, which was restricted to patients untreated with any adjuvant chemotherapy.

Univariate and multivariate analyses in MFS in patients untreated with systemic adjuvant chemotherapy

Conclusion. These data demonstrate overexpression of PARP1 in a large number of breast cancers and its association with reduced metastasis-free survival. These results further support the development of PARP inhibitors in basal subtype but also potentially in other breast cancer subtypes.