Search

Background: The incidence of breast cancer (BC) increases with age. However, there appears to be a trend that elderly patients with BC receive less than the standard treatment compared to younger ones and this leads to poorer outcomes. The aim of this work was to assess whether chemotherapy has a differential impact on the health related quality of life (HRQOL) of elderly patients with BC, compared to younger ones. Methods: CANTO (NCT 01993498) is a multicenter prospective observational study (26 centers in France) whose main objective is to study the chronic toxicities of anticancer treatments in 12,000 women with localized stages I to III BC. HRQOL was assessed at baseline and at the end of acute treatments (surgery +/- radiotherapy +/- chemotherapy) using the EORTC-QLQ-C30 questionnaire and its BC specific module, the EORTC-QLQ-BR23. HRQOL at the end of acute treatments of patients aged 70 and over treated by chemotherapy was compared to that of the youngest ones. Multivariable mixed linear model was used to assess the determinants of HRQOL for the dimensions with minimal clinically important differences; defined as a mean difference of at least 5 points in HRQOL scores; between younger and older patients. Results: 3079 patients were included in the CANTO cohort from March 2012 to December 2014 and treated by chemotherapy. The mean age at diagnosis was 53 years and 7.3% of this population was ≥70 years old. Older and younger patients treated by chemotherapy were similar for the disease stage at diagnosis (p=0.7869), the tumor histology (p=0.07), the SBR grade (p=0.2350) and the HER2 status (p=0.4268). However, elderly patients treated by chemotherapy were more likely to be overweight (p Citation Format: Sandrine Dabakuyo Yonli, Oumar Billa, Olivia Dialla, Florence Joly, Sarah Dauchy, Charles Cecile, Cedric Lemogne, Patrick Soulie, Marie-Ange Mouret-Reynier, Carole Tarpin, Mario Campone, Sophie Guillermet, Jean-Marc Ferrero, Sylvie Giacchetti, Florence Lerebours, Suzette Delaloge, Anne-Laure Martin, Sibille Everhard, Charles Coutant, Fabrice Andre, Patrick Arveux. Impact of chemotherapy on the health related quality of life among elderly patients with localized breast cancer: Findings from the prospective CANTO cohort [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P6-11-07.

Rationale: while large efforts have been done to characterize early breast cancer, little is known about the genomic landscape of metastatic breast cancer. In the present study, we performed whole exome sequencing of 800 metastatic breast cancers, in order to identify new candidate targets and better stratify patients eligible for innovative therapies. Patients and Methods: Patients were selected to present a metastatic breast cancer and to have received a biopsy in the context of precision medicine trials (SAFIR01, SAFIR02, PERMED, MOSCATO, SHIVA). Samples with >30% cancer cells, and normal DNA, were sequenced using Hiseq and Novaseq. Drivers were identified using MutSigCV. Actionability of somatic genetic alterations was determined based on OncoKB. Decomposition of mutational signatures was performed using deconstructSigs. Prognostic value was assessed using a cox model. TCGA database was used as comparator to identify gene alterations enriched in metastatic samples. Results: results presented in the current abstract are based on the first 629 patients analyzed.Sequencing was performed in 387 patients with HR+/Her2- breast cancer, 186 triple negative breast cancers, and 32 Her2-overexpressing breast cancers. only 9 patients received a pretreatment with a CDK4 inhibitor. 24 driver genes were significantly mutated. In patients with HR+/Her2- breast cancer, 11 genes were found more frequently mutated in the metastatic setting as compared to early stage breast cancer. This includes TP53 (29%), KMT2C (13%), NCOR1 (8%), NF1 (7%), RB1 (4%), C16orf3 (2%), FRG1 (6%), ESR1 (21%), RIC8A (4%), AKT1 (7%), PLSCR5 (2%). In addition, in the whole population, KRAS was found mutated in 3% of samples (G12A/C/R/V) while its frequency of mutation in early breast cancer is We further assessed the mutational signatures in order to better understand which mutational processes could drive cancer progression. Metastatic HR+/Her2- mBC presented an increase in APOBEC, S3 (HRD), S10 (POLE-associated signature), S17 signatures as compared to early HR+/Her2- BC. Conclusion: the present study, based on 629 patients, identifies 11 driver gene alterations and four mutational processes enriched in HR+/Her2- metastatic breast cancers. Final results on 800 patients will be presented. Citation Format: Andre F, Filleron T, Ng C, Bertucci F, Letourneau C, Jacquet A, Piscuoglio S, Jimenez M, Bachelot T. Genomic characterisation of metastatic breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr GS1-08.

Background: Palbociclib (PAL) in combination with endocrine therapy (ET) is a preferred treatment option for untreated and previously treated patients with hormone receptor–positive (HR+), human epidermal growth factor receptor 2 (HER2)–negative ABC. Patients with bone only disease could derive prolong disease control with single agent ET. Further understanding the predictive markers for sensitivity to ET alone and P+ET in both endocrine sensitive and resistant settings could help guide clinical treatment sequencing decision. We performed biomarker (BM) analysis of baseline tumor tissues in both PALOMA-2 (P2) and PALOMA-3 (P3) in bone only (BO) vs non-bone (NBO) only patients. Methods: Postmenopausal women (N=666) with no prior systemic therapy for ER+/HER2-ABC were randomized 2:1 to receive PAL+LET or matching placebo (PBO)+LET in P2. In P3 women (N=521) who progressed on prior endocrine therapy were randomized 2:1 to receive PAL+ FUL vs matching PBO+FUL. All pts including those with BO disease had to consent to provide tumor tissues to participate the current study. EdgeSeq Oncology Biomarker Panel (HTG Molecular Diagnostics; Tucson, AZ) was used for mRNA profiling. Cox regression analysis was used to evaluate the association of BO disease with treatment effect as well as the association of gene expression with treatment effect in BO vs NBO pts. Results: Treatment effect in the BM population from each study was consistent with that observed in the respective ITT population. In ITT population, BO disease was reported in 151 (23%) pts of P2 with mPFS 36.2mo PAL+LET arm (n=103) vs 11.2mo LET arm (n=48), HR=0.40 (0.26-0.62) and in 124 (24%) pts of P3 with mPFS 14.3mo PAL+FUL arm (n=86) vs 9.2mo FUL arm (n=38), HR=0.64(0.38-1.06). BO disease was reported in 107 (24%) pts in P2 and 81 (27%) pts in P3 BM population. In P2, longer mPFS was seen with PAL+ LET vs PBO+LET in BO vs NBO disease (30.6 mo vs 11.2 mo; HR=0.42 (0.25-0.69) in BO pts and 22.1 mo vs 13.8 mo HR=0.67 (0.52-0.88) in NBO pts) although the interaction effect was not statistically significant. Similar results were also observed in P3 with longer mPFS in PAL+FUL vs PBO+FUL in BO vs NBO disease (16.6 mo vs 11.2 mo HR=0.79 (0.40-1.56) in BO pts and 11.1mo vs 3.5 mo HR=0.47 (0.34-0.66) in NBO pts) with insignificant interaction effect. In both studies, BO pts had higher incidence of luminal A disease (BO vs NBO: P2=62% vs 47%; P3=54% vs 40%) and lower rate of luminal B disease (P2=19% vs 33%; P3: 26% vs 33%). In both studies, baseline ESR1, Cyclin E and CDK 4 gene expression level were comparable between BO vs NBO pts. Elevated CDK4 gene expression level was associated with resistance to LET in P2 and lower Cyclin E gene expression predicts the effect of PAL in P3. Conclusion: mPFS was significantly prolonged in BO pts with the addition of PAL to ET. BO disease was more frequently associated with luminal A subtype in ABC setting. CDK4 expression level was similar to NBO disease pts but associated with resistance to single agent LET. Citation Format: Finn RS, Turner NC, Liu Y, Rugo HS, Loibl S, Diéras V, Slamon DJ, André F, Gelmon K, DeMichele A, Loi S, Zhang Z, Giorgetti C, Gauthier E, Huang Bartlett C, Cristofanilli M. Biomarker analysis of CDK 4/6 and endocrine pathways in hormone-receptor positive (HR+) advanced breast cancer (ABC) bone only disease patients: A joint analysis of PALOMA-2 and PALOMA-3 studies [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-18-03.

Background: Hyperactivation of the phosphatidylinositol-3-kinase (PI3K) pathway can occur due to PIK3CA mutations, present in ~40% of patients (pts) with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2—) ABC. SOLAR-1, a Phase 3 randomized, double-blind trial (NCT02437318), investigated efficacy and safety of ALP (α-specific PI3K inhibitor) + FUL in pts with HR+, HER2— ABC. ALP+FUL met the primary endpoint by significantly extending progression-free survival (PFS) vs placebo (PBO) + FUL in the PIK3CA-mutant cohort (hazard ratio [HR] 0.65; 95% CI 0.50—0.85; p=0.00065; median 11.0 vs 5.7 months). Here we report overall survival (OS), subgroup data and safety in the PIK3CA-mutant cohort, and PFS by circulating tumor (ct)DNA PIK3CA mutation status in the total population. Methods: Enrollment was open to men/postmenopausal women with PIK3CA-mutant HR+, HER2— ABC and 1 prior line of endocrine therapy. Pts were randomized (1:1) to ALP (300mg/day) + FUL (500mg every 28 days and Cycle 1 Day 15) or PBO+FUL. OS was the key secondary endpoint. PFS was analyzed by PIK3CA mutant status in ctDNA, and in important prognostic subgroups, including line of treatment in ABC and prior cyclin-dependent kinase 4/6 inhibitor (CDK4/6i) use. Safety was analyzed in the PIK3CA-mutant cohort. Results: 341 pts in the PIK3CA-mutant cohort received ALP+FUL (n=169) or PBO+FUL (n=172). Median follow-up from randomization to data cut-off was 20.0 months. At data cut-off, 92 deaths had occurred (52% of the total 178 pts planned for final OS analysis); 40 for ALP+FUL (24%) and 52 for PBO+FUL (30%). OS results were immature at data cut-off (HR 0.73; 95% CI 0.48—1.10; p=0.06; median not estimable vs 26.9 months). There was a 45% risk reduction in PFS for pts with ctDNA PIK3CA mutations (HR 0.55; 95% CI 0.39—0.79; n=186); 20% for pts without (HR 0.80; 95% CI 0.60—1.06; n=363). PFS treatment effect for ALP+FUL vs PBO+FUL was generally consistent across subgroups of interest, with a risk reduction of 29% for pts receiving first-line (1L) treatment (HR 0.71; 95% CI 0.49—1.03; n=177), and 39% for 2L treatment (HR 0.61; 95% CI 0.42—0.89; n=161); 52% in pts with prior CDK4/6i (HR 0.48; 95% CI 0.17—1.36; n=20) and 33% in pts without (HR 0.67; 95% CI 0.51—0.87; n=321). Most frequent all-grade adverse events (AEs; ≥40% in either arm by single preferred term; ALP+FUL vs PBO+FUL) were hyperglycemia (65% vs 9%), diarrhea (54% vs 11%), nausea (46% vs 20%), and rash (40% vs 6%). Grade 3/4 AEs in ≥10% pts in either arm were hyperglycemia (fasting plasma glucose >250mg/dL; 37% for ALP+FUL vs Conclusions: ALP+FUL showed consistent clinically meaningful treatment benefit for pts with ctDNA PIK3CA mutant status, and across pt subgroups, including pts with/without prior treatment for ABC and prior CDK4/6i use. OS data were not yet mature at the data cut-off, but OS appeared numerically longer for ALP+FUL vs PBO+FUL after 52% of events. Key words: advanced breast cancer; PI3K; alpelisib; endocrine therapy Citation Format: Juric D, Ciruelos E, Rubovszky G, Campone M, Loibl S, Rugo HS, Iwata H, Conte P, Mayer IA, Kaufman B, Yamashita T, Lu Y-S, Inoue K, Takahashi M, Pápai Z, Longin A-S, Mills D, Wilke C, Sellami D, Andre F. Alpelisib + fulvestrant for advanced breast cancer: Subgroup analyses from the phase III SOLAR-1 trial [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr GS3-08.

Background Smoothened (SMO) inhibitors, blocking the sonic hedgehog pathway, have been approved for advanced basal cell carcinoma (aBCC). Safety analyses reveal a high rate of adverse events (AEs) and, most of the time, vismodegib is most commonly stopped when the best overall response is reached. The long-term evolution of aBCC after vismodegib discontinuation is poorly described. The aim of this study is to evaluate the efficacy and safety of the SMO inhibitors (SMOis) available (vismodegib and sonidegib) following rechallenge after complete response (CR) following an initial treatment by vismodegib. Materials and methods This real-life, retrospective, multicenter and descriptive study is based on an extraction from the CARADERM accredited database, including 40 French regional hospitals, of patients requiring BCC systemic treatment. Results Of 303 patients treated with vismodegib, 110 achieved an initial CR. The vast majority of these patients (98.2%) stopped vismodegib, notably due to poorly tolerated AEs. The CARADERM database provided a median follow-up of 21 months (13.5-36.0 months) after CR. Of the 110 patients, 48.1% relapsed after a median relapse-free survival of 24 months (13.0-38.0 months). Among them, 35 patients were retreated by an SMOi and the overall response rate was 65.7% (34.3% of CR and 31.4% of partial response). The median duration of retreatment was 6.0 months (4.0-9.5 months). Conclusion Our real-life study, carried out on patients with complex clinical pictures, shows that after treatment discontinuation, 48.1% of patients achieved CR relapse within an average of 24 months (13.0-38.0 months). It emphasized that even though rechallenge can be considered as a therapeutic option, efficacy seems to decrease, suggesting the development of resistance mechanisms., Highlights • Real-world experience of SMOis in a French national multicenter and retrospective study. • Efficacy and tolerance detailed in rechallenge, with long-term evolution. • Focus on a population with Gorlin syndrome.

Background Preclinical and clinical data suggest that HER2-positive (HER2+) breast cancer (BC) will be amendable to immunotherapeutic approaches. We evaluated pembrolizumab with trastuzumab in patients (pts) with trastuzumab-resistant HER2+, PD-L1 positive (PD-L1pos), unresectable loco-regional or metastatic BC and a parallel cohort of pts with HER2+, PD-L1 negative (PD-L1neg) BC during the phase II study. Methods: Pts with advanced BC and progression on prior trastuzumab-based therapies, ECOG 0-1, and a metastatic tumor biopsy in the last year were eligible. HER2 positivity and quantity of tumor-infiltrating lymphocytes (TILs) on H&E slide were centrally evaluated. PD-L1 score was assessed by Merck central lab. Tumor imaging was performed at weeks 12, 18, 24 and every 12 weeks, thereafter. Primary endpoints were safety of the combination (phase Ib) and objective response rate (ORR) per RECIST 1.1 (phase II). Secondary endpoints were PFS, duration of response, and OS. Phase Ib was a 3+3 dose-escalation of 2 pembrolizumab doses (2mg/kg, 10mg/kg) Q3W. In phase II, pts received pembrolizumab 200mg Q3W for 24 months or until disease progression. Clinically stable pts with progression were allowed to continue pembrolizumab until confirmation on subsequent assessment. Pts with isolated CNS progression were also allowed to continue pembrolizumab after local treatment. Planned total enrollment was 61 pts. For the phase II PD-L1pos cohort, a Simon two-stage design (N=40; proceed if ≥2/17 respond) was used which had 85% power to compare ORR of 7% vs. 22% (1-sided α=0.05). For the PD-L1neg cohort, a single-stage design with 15 pts had >95% power to compare ORR of 1% vs. 20% (1-sided α=0.14). Clinicaltrials.gov: NCT02129556. Results: 6 pts enrolled in phase Ib between April and July 2015; no DLTs were observed. The PD-L1pos cohort enrolled 40 pts between August 2015 and September 2016. The PD-L1neg cohort enrolled May 2016 to April 2017, stopping after 12 pts due to low rate of PD-L1 negativity, maintaining >90% power to detect the target difference in ORR. PD-L1 testing labs changed in April 2016. Prior to this time, QualTek PD-L1 positive was defined as ≥1% on tumor or TILs. Using the Dako 22C3 antibody, positive was defined as tumor PD-L1 combined positive score (CPS)≥1%. 146 pts were screened to enroll 58 pts. Of screened pts, median stromal TILs was 1% (mean: 4.8%, SD: 9.1%, range: 0 to 60%; n=127); 52% of pts were PD-L1pos, with higher positivity rates while using the Dako assay compared with Qualtek (65% vs. 43%, p=0.009). Median TILs of pts in the PD-L1pos cohort was 2% (mean: 8.1%, SD: 11.2%, range: 0 to 40%) and 0% (mean: 1.2%, SD: 2.2%, range: 0 to 5%) in the PD-L1neg cohort. Of enrolled pts, median age was 51yrs (range: 28-72), 69% had visceral metastases. 29% of pts received prior pertuzumab, 72% had prior T-DM1, 40% prior lapatinib. 38% of pts were ER-positive, 62% were ER-negative. Median TILs in enrolled ER pos and ER neg pts were 1.5% and 2.0%, respectively. PD-L1 positivity rates were also not significantly different by ER status (p=0.5). Final safety data and efficacy results will be presented at the meeting. Citation Format: Loi S, Giobbe-Hurder A, Gombos A, Bachelot T, Hui R, Curigliano G, Campone M, Biganzoli L, Bonnefoi H, Jerusalem G, Bartsch R, Rabaglio-Poretti M, Kammler R, Maibach R, Smyth MJ, Di Leo A, Colleoni M, Viale G, Regan MM, Andre F. Phase Ib/II study evaluating safety and efficacy of pembrolizumab and trastuzumab in patients with trastuzumab-resistant HER2-positive metastatic breast cancer: Results from the PANACEA (IBCSG 45-13/BIG 4-13/KEYNOTE-014) study [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 GS2-06.

FGFR1 amplification could mediate resistance to endocrine therapy and FGFR1 inhibition reverses this resistance. This phase Ib seeks to evaluate whether the combination of lucitanib, a potent FGFR/VEGFR/PDFGR inhibitor, in combination with fulvestrant, an endocrine agent, reverses resistance to fulvestrant. Eligible patients for this study were postmenopausal with ER+/HER2- mBC and have relapsed during or after treatment with fulvestrant. There were 2 parts in the study: a dose allocation to assess the tolerability of the combination in terms of DLTs and MTD using a modified Continual Reassessment Method (mCRM) [part I] and a dose expansion, with patients assigned to 2 different cohorts based on FGFR amplification, to further evaluate the tolerability of the combination and to identify the recommended phase II dose (RP2D) [part II]. Surrogate target hitting biomarkers were also dosed at baseline and on-treatment. The sponsor decided to halt the clinical development in mBC indication and the study was prematurely terminated after 18 patients (15 in part I and 3 in part II). The presentation will focus on these 18 patients. Patients had ECOG PS 0 or 1 and median number of previous treatments in metastatic setting was 3. Two doses of lucitanib (10mg daily n=9 and 12.5mg daily n=6) in combination with 500 mg/month of fulvestrant were tested in part I. At the 10mg dose level, one patient experienced a DLT (grade 3 hypertension). Based on global lucitanib development program data, it was decided to start Part II with lucitanib 10mg daily. The most common related grade ≥3 toxicities occurring in more than 10% of patients were hypertension (78%) and asthenia (22%). All patients required at least one dose interruption mainly for toxicities, while 13 patients (72%) required at least a dose reduction for toxicities. Thirteen patients (72%) withdrew from the study for disease progression, 3 (17%) for adverse events (at 10mg) and 2 (11%) for non-medical reasons. Three patients achieved a confirmed partial response (as per RECIST v1.1), one at 10mg and two at 12.5mg. About 55% of the patients experienced clinical benefit with a median duration of the benefit of 39.6 weeks and a maximun duration of the benefit of 79.1 weeks for 1 patient (PR at Cycle 4). Biomarker modulations were consistent with lucitanib mode of action; targeting VEGFRs (significant increase of VEGFA, IL8, PlGF) and FGFR1 (significant increase of FGF23). The combination is feasible but requires close patient monitoring and intensive management of adverse events. Those are in line with the anti-angiogenic activity of lucitanib. 10mg (N=12)12.5mg (N=6)All (N=18)Objective Response Rate (ORR)n(%) 11 (8.3)2 (33.3)3 (16.7) 95% CI 3[1.5;35.4][9.7;70.0][5.8;39.2]Clinical Benefit Rate (CBR)n(%) 24 (33.3)6 (100.0)10 (55.6) 95% CI 3[13.8;61.0][61.0;100.0][33.7;75.4]Duration of Clinical Benefitmedian (weeks)28.171.339.6 95% CI 3[27.9; 32.7][29.1; 79.1][27.9; 79.1]1: CR or PR 2: CR or PR or stabilization (SD or NonCR/NonPD) >24 weeks or at end of cycle 6 3: 95% Wilson method of Confidence interval of the estimate Citation Format: Campone M, Bachelot T, Penault-Llorca F, Pallis A, Agrapart V, Pierrat M-J, Poirot C, Paux G, Dubois F, Xuereb L, Robert R, Andre F. A phase Ib study of oral administration of lucitanib in combination with fulvestrant in patients with HR+ metastatic breast cancer (mBC) [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 P1-09-11.

Purpose: There is a need to identify which patients present a high risk of relapse after optimal adjuvant therapies, in order to better define which population of early breast cancer patients is eligible to new drugs. Previous studies have shown that gene copy numbers drive breast cancer progression. In the BIG1-98 trial, FGFR1 (8p11) and ZNF217 (20q13) amplifications were associated with poor outcome in patients with HR+/Her2- early breast cancers. In the present study, we aim to validate these findings using retrospective analysis of samples prospectively collected in a randomized trial (PACS04). Patients and methods: Tumor DNA was extracted from FFPE samples obtained from patients included in PACS04 trial. This trial compared FEC to ED75 in patients with node positive, early breast cancer and included 3000 patients between 2001 and 2004. The biomarker study focuses on HR+/Her2- breast cancer. 500 samples were collected and hybridization was performed on 390. DNA copy numbers were assessed on 900 genes using Oncoscan FFPE assay kit. Prognostic value of FGFR1 (8p11) and ZNF217 (20q13) copy numbers was assessed in a Cox model that included prognostic parameters. Prognostic value was assessed using CNA as continuous variable, and with a predefined cut-off (CN>=3). Metastases-free survival (MFS) was chosen as endpoint for this biomarker study. Results: Tumor copy numbers alterations were obtained in 390 patients with HR+/Her2-, node positive, early breast cancer. 31% of the patients presented more than 3 lymph node involved, and 25% a poorly differenciated tumor. The median tumor size was 20 mm. 120 (30%) and 112 (28%) patients presented FGFR1 and/or ZNF217 gene gain with a cut-off for amplification predefined at 3 copies. When assessed as continuous variable, FGFR1 (HR: 1.02, 95%CI:1.007-1.04, p=0.0045) and ZNF217 (HR: 1.011, 95%CI:1.003-1.01, p=0.006) copy numbers were associated with MFS. The 10 years MFS rates were 68% (95%CI: 59-78%) and 85% (95%CI: 81-91%) in patients with FGFR1-gained and FGFR1-normal tumors respectively (HR: 2.51, 95%CI:1.56-4.01, p=0.0001). The 10 years MFS rates were 72% (95%CI:65%-83%) and 83% (95%CI: 77-87%) in patients with ZNF217-gained and ZNF217-normal tumors (HR: 1.79, 95%CI:1.12-2.86, p=0.013). In a multivariate analysis, FGFR1-gain (HR: 2.45, 95%CI:1.42-4.22, p=0.0012) and ZNF217-gain (HR:1.78, 95%CI:1.00-3.17, p=0.049) were associated with a higher likelihood of metastases and/or death. The 10 MFS rate was 88% (95%CI: 83-94%) for patients who presented FGFR1- and ZNF217-normal tumor (n=191), and 71% (95%CI: 66-76%) in patients presenting FGFR1 and/or ZNF217 gene gain. Conclusion: Using samples prospectively collected in a randomized trial, this study validates the prognostic value of FGFR1- and ZNF217- copy numbers in patients who present HR+/Her2- early breast cancer. Exploratory analyses on copy number alterations, together with targeted sequencing are ongoing and will be presented. Citation Format: Andre F, Penault Llorca F, Carene D, Roche H, Delaloge S, Lacroix L, Scott V, Richon C, Lacroix-Ticrit M, Lemonnier J, Michiels S. FGFR1 / ZNF217 copy numbers and outcome in patients with node positive, HR+/Her2- early breast cancer: A genomic analysis of PACS04 trial [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 PD4-10.

Background: to determine the distribution and evolution of mutational processes in metastatic breast cancers (mBC), together with their clinical relevance Methods: Whole exome sequencing (Hi-Seq, Illumina) and determination of copy number alterations (CNA) (CGH array / SNP6.0) were performed in 240 and 692 metastatic breast cancers respectively. Mutational processes were defined according to Alexandrov (Nature, 2013). Homologous Recombination Deficiency (HRD) was determined by genome wide assessment of loss-of-heteroygosity (LOH) on SNP6.0 (n = 210). Finally, genomic instability was assessed by the % of genome altered assessed by CGH / SNP6.0 Results: Whole exome sequencing showed that HR+/Her2- metastatic breast cancer presented an increased contribution of APOBEC-related signatures, as compared to early breast cancer (TCGA) (58% of the mutations vs 31%, p < 0.0001). Twelve percent of the HR+/Her2- mBC acquired an hypermutator genotype ( > 200 non-synonymous mutations). This acquisition of an hypermutator genotype was confirmed in five paired primary-metastatic samples. An operational APOBEC-related signature 13 was associated with a poor outcome in a multivariate analysis (HR: 1.75, 95%CI: 1.1-2.7, p = 0.017). High LOH score (HRD) was observed in 30% of HR+/Her2- mBC as compared to 13% of early HR+/Her2- early BC (p < 0.0001). The opposite was observed in TNBC (43% in mTNBC versus 58% in early TNBC ,p = 0.032). High LOH score was associated with a trend for poor outcome in HR+/Her2- mBC (multivariate 1.67, 95%CI: 0.949-2.951, p = 0.075).The % of genome altered was associated with a poor outcome in multivariate analyses both in the overall and HR+/Her2- mBC (HR / 10 increase:1.144, 95%CI:1.038-1.261, p = 0.007 and HR:1.18, 95%CI:1.037-1.344, p = 0.012 respectively). Copy number analyses identified 143 genes that are more frequently amplified as compared to early breast cancers (FDR < 0.01) Conclusions: metastatic HR+/Her2- metastatic breast cancer present an increased in APOBEC-related mutational burden and in LOH score as compared to early breast cancers. APOBEC-related signature 13 and genome instability are associated with a poor outcome and could be used in the future to better stratify metastatic breast cancer patients. Citation Format: Patsouris A, Filleron T, Jacquet A, Goncalves A, Bonnefoi H, Letourneau C, Bachelot T, Jimenez M, Andre F. Mutational processes, genome evolution and outcome in metastatic breast cancers [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 PD8-12.

Background A genomic-driven therapeutic strategy in metastatic breast cancer (MBC) was recently demonstrated as feasible in the clinical practice, but its actual impact on patient outcome remains elusive. SAFIR02 study is an ongoing national multicentric phase II randomized trial evaluating targeted therapies matching specific genomic alterations (GA) administered as maintenance after objective response and/or stable disease obtained with chemotherapy in HER2-negative MBC patients. This analysis reports on feasibility of the procedure and the rate of identified actionable targets. Methods Eligible MBC patients (PS=0/1, first- or second-line of chemotherapy, HER2-negative/hormone receptor (HR)-negative or endocrine resistant HR-positive; measurable per RECIST 1.1; accessible to tumor biopsy; no bone metastases-only disease, no major organ dysfunction) were subjected to tumor biopsy for genomic analysis (CGH arrays, Affymetrix Cytoscan; NGS, Ion Torrent PGM, AmpliSeq, panel of around 50 genes). Actionable GA were identified and corresponding targeted therapies were proposed by a multidisciplinary tumor board (MTB). Patients received cytotoxic-based treatment at physician's choice and those with stable or responding disease after 6 to 8 cycles (or at least 4 if stopped for toxicity reason) and targetable GA, were offered randomization between targeted therapy or chemotherapy maintenance until progression or intolerance (main study). Since January 2016, an amendment was made to propose to patients without targetable alteration a randomization between anti-PD-L1 (MEDI4736) or standard chemotherapy maintenance (substudy). Results Between March 2014 and May 2016, 457 patients have been enrolled at 21 centers. Genomic analyses could not be obtained in 107 cases (23%) due to either biopsy failure (n= 40; 9%) or low cellularity (n=67; 14%). Of the 307 patients reviewed by the MTB, 197 (64%) had an actionable GA, including PIK3CA-PIK3CB-PIK3R1 (n=51), FGF4 or FGFR1/2 (n= 42), BRCA1/2 (n=15), AKT1/2/3 (n=13), BRAF/KRAS/NRAS (n=13), HER2/3 (n=10), NF1-FRS2 (n=10), MTOR-RPTOR-TSC2 (n=8), PTEN (n=7), STK11 (n=7), IGF1R (n=7), EGFR (n=5). Therapeutic proposals by MTB included AZD5363 (n=71), AZD4547 (n=42), AZD2014 (n=23), selumetinib (n=23), olaparib (n=16), AZD8931 (n=15), vandetanib (n=5), bicalutamide (n=2). In an exploratory analysis involving 157 patients, the rate of targeted therapy proposal by MTB markedly differed between triple-negative patients (TNBC; 24 of 48, 50%) and HER2-negative/HR-positive patients (92 of 109, 84%; p=6.14. 10-6, Chi-2 test). At the time of the analysis, 85 patients have been randomized (main study, 68; substudy, 17). Causes of randomization failure (n=108) included disease progression (n=45) or death (n=25), non-eligibility criteria (n=27), patient/physician's decision (n=11). Conclusion A large number of patients had identified targetable GA. Of note, the rate of targeted therapeutic proposal was significantly lower in TNBC than in HER2-negative/HR-positive patients. Rapidly progressing disease may impede ultimate randomization. Citation Format: Gonçalves A, Bachelot T, Lusque A, Arnedos M, Campone M, Bièche I, Lacroix L, Pierron G, Dalenc F, Filleron T, Sablin M-P, Jimenez M, Ferrero J-M, Lefeuvre-Plesse C, Bonnefoi H, Attignon V, Soubeyran I, Jezequel P, Commo F, André F. High-throughput genome analysis and therapeutic decision for patients with HER2-negative metastatic breast cancer: First feasibility and molecular results of the randomized phase II study SAFIR02 BREAST (UCBG-0105/1304) [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr PD1-08.

Background:Although ER+/HER2– breast cancer has a better overall prognosis than other breast cancer subtypes, there is a high-risk subpopulation characterized by high-grade tumors and decreased sensitivity to endocrine therapy, higher responsiveness to chemotherapy and worse prognosis. A large meta-analysis of prospective studies focusing on neoadjuvant chemotherapy (NAC) for treatment of stage I-III breast cancer demonstrated that increased pathologic complete response (pCR) rates at surgery were associated with improved survival. This correlation was observed across triple-negative breast cancer (TNBC), HER2+ breast cancer, and high-grade HR+/HER2- breast cancer. Specifically, patients with a pCR after NAC had a 5-year event-free survival (EFS) rate of 90%, whereas patients who did not achieve a pCR had a 5-year EFS rate of 60%.Therefore, increasing pCR rates after NAC may have a substantial impact for patients with high-risk early-stage HR+/HER2– breast cancer. KEYNOTE-756 is a global, randomized, double-blind, phase III study of pembrolizumab (vs placebo) + chemotherapy as neoadjuvant treatment, followed by pembrolizumab (vs placebo) plus endocrine therapy as adjuvant treatment for patients with high-risk, early-stage ER+/HER2– breast cancer. Methods: Patients with T1c-2 cN1-2 or T3-4 cN0-2 grade 3 or grade 2 with Ki-67 ≥30%, invasive, ductal ER+/HER2– breast cancerwill be stratified by lymph node involvement (positive vs negative), tumor PD-L1 status (positive vs negative), ER positivity (≥10% vs Citation Format: Cardoso F, Bardia A, Andre F, Cescon DW, McArthur H, Telli M, Loi S, Cortés J, Schmid P, Harbeck N, Denkert C, Jackisch C, Jia L, Tryfonidis K, Karantza V. KEYNOTE-756: A randomized, double-blind, phase III study of pembrolizumab versus placebo in combination with neoadjuvant chemotherapy and adjuvant endocrine therapy for high-risk early-stage ER+/HER2– breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr OT3-04-03.

Background: A subset of patients with HER2-positive metastatic (mBC) or locally advanced breast cancer (aBC) respond to 1st line trastuzumab-based regimen and have no disease progression for several years. These long-term responders were evaluated in daily practice in the LORHA study. Patients and disease characteristics were presented earlier (ESMO 2012). Material and Methods: Lorha is an ambispective, multicenter, non-interventional study conducted in 57 centers in France. Eligible patients were women aged more than 18 years with HER2-positive mBC or aBC treated with 1st line trastuzumab and progression-free for at least 3 years after starting trastuzumab. The primary objective was to describe clinical and tumor characteristics. Secondary objectives included Progression Free Survival (PFS), Overall Survival (OS), treatment administration and safety. Patients were followed-up for 12 months. Final results are presented here. Results: Overall 160 patients were enrolled; 128 were eligible for efficacy analysis and 154 for safety analysis. Median age was 61 years [34-95]. Tumor characteristics were: invasive ductal carcinoma for 112 patients (88%), positive hormonal receptors in 72 patients (58%). At initial diagnosis, 91 patients (71%) had stage I - II tumors and 37 patients (29%) had de novo mBC or aBC. The main metastatic localizations were bone, liver, lymph nodes and lung in 54 (43%), 44 (35%), 25 (20%) and 24 (19%) patients, respectively. Median 1st line trastuzumab treatment duration was 4.5 years [0.8 - 12.1]. Trastuzumab 1st line was associated with a taxane-based chemotherapy in 104/128 patients (81%). Median PFS was 6.4 years [5.7; - Not reached] and median overall survival was not reached. At 1-year follow-up, 23 deaths (15% of patients) were reported, mainly due to disease progression (21 patients - 91%). No toxic death related to trastuzumab was observed. Grade 3-4 adverse events reported prospectively occurred in 14 patients (10%) and included two patients (1.2%) with grade 3-4 cardiac events related to trastuzumab (congestive cardiac failure and left ventricular dysfunction), both events resolved, one with sequelae. Conclusions: Final results of the LORHA study show that long responding patients receiving first-line trastuzumab-based treatment for aBC or mBC in daily practice, can reach a median PFS of more than 6 years, combined with an acceptable safety profile. Median overall survival is still not reached. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-16-08.

Background Despite a growing number of studies exploring the underlying mechanisms, resistance to trastuzumab (TTZ) in HER2+ breast cancer (BC) remains elusive and an important issue in clinical practice. Among the proposed mechanisms, involvement of the PI3K/Akt/mTOR pathway is one of the best characterized. In a previous preclinical and clinical study (n = 54) (Bergé Y et al, SABCS 2009 and [1]), we have shown a significant correlation between eIF4E (a downstream effector of the PI3K/Akt/mTOR pathway) expression level as determined by immunohistochemistry (IHC) on initial tumor biopsy and pathological complete response (pCR) in patients with HER2+ BC treated with a neoadjuvant TTZ-containing regimen. The aim of this study was to validate these findings on an independent and larger cohort of HER2+ BC treated in neoadjuvant setting. Method In this multicenter (n = 9) study, 274 patients (pts) with HER2+ BC treated were included. All pts received neoadjuvant chemotherapy regimen containing TTZ, consisting mostly of 3-4 cycles of FEC (5FU, Epirubicin, Cyclophosphamid) followed by 3-4 cycles of TTZ–Docetaxel (n = 116). 74 pts received 6 cycles of TTZ–Docetaxel, 18 received 6 cycles of TTZ–Docetaxel-Cyclophosphamid, 12 received 6 cycles of TTZ–Docetaxel-carbolpatine, and the remaining 54 pts received other regimens. Pathological response was assessed according to Sataloff and Chevallier criteria. Unstained slides from the initial tumor biopsy were centrally collected for biomarkers analysis. Expression levels of eIF4E, p-4EBP1 and pS6 were determined by IHC. An immunoreactive score (IRS) combining the percentage of stained tumor cells and staining intensity was assessed by two pathologists. Results Median age at diagnosis was 50 years [range 22–84]. Most pts presented with T2 (52.6%), N1 (53.9%) stage, with a median tumor size of 40mm [range 0-150mm] as determined on clinical examination. Tumors were mainly invasive ductal of no special type (93.8%), of histological grade III (50%) and II (45%), ER+ (54.4%). Breast conserving surgery with free margins was achieved in 50.9% of pts. pCR (breast Sataloff TA) was observed in 53% of pts, and in 47.6% when considering both breast and lymph node response. Out of the 274 pts, 257 had sufficient tumor on the initial biopsy for biomarker analysis. Preliminary results showed a median eIF4E IRS of 6 [range 1-12], with 52.1% of cases displaying a low eIF4E expression level (IRS£6). The agreement for inter-observer assessment of eIF4E status was good (k = 0.618, 95% CI [0.523-0.713]). Scoring of p-4EBP1 and pS6 are ongoing and correlations of pCR with eIF4E/4EBP1 axis will be presented. Conclusion In this large multicentre retrospective study, the rate of pCR obtained in neoadjuvant setting of HER2+ BC is similar to those described in the literature. Thorough tumor collection allows biomarkers analysis, which has been specifically focused on downstream effectors of the PI3K/Akt/mTOR pathway. Reference[1] Zindy P, Bergé Y, Allal B et al. Formation of the eIF4F translation-initiation complex determines sensitivity to anticancer drugs targeting the EGFR and HER2 receptors. Cancer Res. 2011;71(12):4068-73. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-14-13.

Background: PI3-kinase (PI3K) and MAP kinase (MAPK) are the main signaling pathways implicated in molecular oncogenesis. In breast cancer, inhibition of these signaling pathways has been largely envisaged by means of targeted therapy. Unicancer RADHER study aimed at evaluating the efficacy of adding everolimus (E) to trastuzumab (T) as preoperative therapy for primary HER2+ operable breast cancer patients and to evaluate molecular response biomarkers. We report here the investigation of the expression of phosphoproteins from PI3K and MAPK signaling pathways as predictive biomarkers of clinical and pathological response as well as pharmacodynamic markers of treatment activity. Methods: 82 eligible patients were randomized to receive T alone (loading dose 4 mg/kg, then 2 mg/kg/week), or T+E (10 mg/day) for a 6-week pre-operative treatment. Clinical response rate (cRR) was determined from OMS criteria with complete and partial responses being considered as “ responders “ and stable and progressive diseases as “ non responders “. Pathological response rate was evaluated according to Sataloff classification, with Ta and Tb being considered as “ responders “ and, Tc and Td as “ non responders “. The expression levels of phosphorylated-AKT (p-AKT), p-GSK3b, p-S6 kinase, p-MEK1, p-ERK1/2, p-P90RSK, p-IGF1R as well as p-P38MAPK were quantitatively assessed using multiplex bead immuno-assay. All patients had baseline needle frozen biopsies taken before initiation of the treatment, at cycle 4 as an option and at surgery. Before being submitted to total protein extraction, all biopsies were validated by a senior pathologist after HE slide examination to ensure a tumor content >50%. 36 pairs associating baseline + surgery tumor specimens and 4 pairs of baseline + cycle 4 biopsies were eligible for protein extraction. Results: No statistically significant relationship was observed between the expression level of any of the phosphoproteins in the initial biopsies and neither the clinical nor the pathological response, overall. After treatment, as compared to the level of expression measured in the initial biopsies, a significant increase of p-GSK3β, p-MEK1, p-ERK1/2, p-P38MAPK was observed in T+E arm and a significant decrease in p-S6 kinase expression in the global patient population. No significant variation was observed in T arm. Additional analysis with immunohistochemistry data is planned and will be presented. Conclusion: In the present study, measuring phosphoproteins expression showed that combining E with T, altered the regulation of signaling proteins from PI3-Kinase and MAP-kinase pathways. No response predictive biomarker could be identified among the phosphoproteins analyzed tending to show that the clinical and pathological response to T and T+E should be driven by additional mechanisms. As a whole, these results validate the use of multiplex bead immuno-analysis for determination of phosphorylated signaling proteins in clinical needle biopsies from breast cancer specimens and its prospective evaluation as biomarker for the activity of targeted therapies. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-27.

Background: Predicting a benefit from trastuzumab in patients with HER2+ breast cancer remains an important goal. Possible mechanisms of resistance include altered receptor antibody interaction, Akt and MAPK pathways, and loss of p27. The objective of this study was to determine the correlation between pMAPkinase (pMAPK), pAKT, p27, IGF-IR protein expression and the benefit of trastuzumab for patients randomized to chemotherapy (CT) alone and CT with trastuzumab. Patients and methods: From May 2004 to October 2007, 120 patients with stage II and III HER2+ breast carcinomas were enrolled in a phase II trial of neoadjuvant chemotherapy (CT) with epirubicin-cyclophosphamide (4 courses) followed by docetaxel ± trastuzumab (T) (4 courses). A complete pathological response (pCR) was defined by the absence of residual invasive carcinoma in the breast and axillary lymph nodes. A tissue microarray was constructed from paraffin-embedded tumor samples collected prior to neoadjuvant chemotherapy. Patients' tumours were scored HER2 3+ immunohistochemically (IHC) or 2+ IHC with HER2 amplification by FISH. Immunohistochemical analysis of pMAPK, pAKT, p27 and IGF-IR was performed on tumor tissue microarrays before CT. The H-score (intensity × %) was evaluated. Specimens were classified as exhibiting high or low expression based on a median value as the cut-off point for each marker. A logistic regression model, including the marker and its interaction with treatment, was used to analyse the markers predictive of a treatment effect on the pCR. The independent predictive value was analysed in a multivariate logistic regression adjusting on the lymph node and ER status. Results: 117/120 (97.5%) patients had sufficient tumor for the analysis. The pCR rate was 19% in the CT arm and 25% in the CT+T arm. The median H-score was: pMAPK = 28, pAKT= 25, p27= 50 and IGF-IR = 15. No significant difference was observed in the pCR rate between the two arms according to pAKT, p27, IGF-IR expression. The pCR rate was higher in CT+T compared to CT alone in patients with high pMAPK expression (OR = 4.7 (0.9–24.2); interaction p = 0.03). No difference was observed in the pCR rate in patients with low pMAPK expression (OR = 0.5 (0.1–1.8). Conclusions: In HER2-positive breast cancers, pMAPK expression evaluated by IHC was significantly associated with a pathological response in the arm with neoadjuvant trastuzumab. High pMAPK expression could be a predictive marker of response to trastuzumab in a CT +T regimen. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P3-06-04.

Introduction: Neoadjuvant chemotherapy increases the likelihood that breast conservation therapy for breast cancer patients will be successful. Breast Cancer Index (BCI), a gene expression assay combining HoxB13/IL17BR ratio and Molecular Grade Index (MGI), is prognostic for the risk of distant recurrence and overall survival in tamoxifen-treated and untreated breast cancer patients. It was previously reported that high risk patients, as determined by BCI, had a 10-fold greater probability of pathologic complete response (pCR) with neoadjuvant chemotherapy than low risk patients. The aim of the current study was to examine the relationship between BCI score and the use of breast conservation surgery (BCS) following treatment with neoadjuvant chemotherapy. Material and Methods: A total of 145 women (tumor size T1, T2 and T3) were treated with neoadjuvant cheomtherapy for stage I-III breast cancer. RNA was extracted from FFPE tumor samples and a real-time RT-PCR assay was completed to generate a BCI score and risk group categorization as previously described (Jerevall et al. Br J Cancer 2011). The relationship between BCS, BCI and clinicopathological factors was examined using univariate and multivariate logistic regression. Results: Of the 145 patients (67% ER+, 54% PR+, 57% >50 y old), 48 (33.1%) underwent BCS. BCI categorized 62 (43%) of patients as low, 50 (34%) as intermediate and 33 (23%) as high risk. The rate of BCS for the three BCI risk categories was 15% (low risk), 48% (intermediate risk) and 45% (high risk). In the low risk group, the rate of BCS was 15% corresponding to a NPV of 85%. This is consistent with previous data from the same cohort, where the NPV of BCI for pCR was 98.4% with only one patient in the low risk group achieving pCR. In univariate analysis, pathological tumor size (pT), ER, PR, grade and BCI were predictors of BCS. A higher BCI score was associated with higher likelihood of BCS (odds ratio of 3.90; CI: 1.45−10.49; p=0.0069). In multivariate analysis, pT and BCI remained significantly associated with BCS, while ER status was not (p=0.23). Results were similar in the subset of patients with T1 and T2 tumors (N=97). In this subset, BCI categorized 42% of patients as low, 37% as intermediate and 21% as high risk and the rate of BCS was 22%, 64% and 60%, respectively. In multivariate analysis of this subset, only BCI was significantly associated with BCS. In all patients, the concordance index based on a model with pT alone was 0.695. When BCI was incorporated into the model with pT, the concordance index increased to 0.801 (p= 0.0002). Conclusion: In this study, we have shown that patients with higher BCI scores were associated with a higher likelihood of receiving BCS after neoadjuvant chemotherapy. Addition of BCI to tumor size increased accuracy in predicting likelihood of BCS. BCI along with standard pathological factors may improve estimation of individual probability of BCS after neoadjuvant chemotherapy. This study gives rise to the hypothesis that patients with low BCI should not be eligible for neoadjuvant chemotherapy since the likelihood of breast conservation is low. Further large confirmatory studies are necessary. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P5-13-08.

Background: Inflammatory breast cancer (IBC) is a relatively rare form of breast cancer. Because this disease exhibits angiogenic properties, bevacizumab (BEV) may improves the efficacy of neoadjuvant chemotherapy (CT) in IBC. Methods: Pts with HER2−negative IBC (IHC 0 or 1+ or FISH/CISH-, T4d, any N) were included to receive 4 cycles of FEC100 + BEV 15 mg/kg, d1 q3w, followed by 4 cycles of Docetaxel 100 mg/m2 + BEV 15 mg/kg q3w. Complete surgery was performed 4–6 weeks after the last dose of CT, followed by radiotherapy and endocrine therapy for hormone receptor positive (HR+) pts. BEV was re-introduced as adjuvant therapy for 10 cycles, giving a total of 18 cycles of BEV. The primary endpoint of the PACS09/Beverly1 trial was to evaluate the pathologic complete response (pCR; Sataloff classification) rate in HER2−negative non-metastatic IBC patients(pts). Results: Between Jan 2009 and Sep 2010, 101 pts were included (100 pts were evaluable). Main baseline characteristics were: median age 49 years; postmenopausal status, 39 pts; SBR 2, 32 pts SBR3 61 pts, both HR-negative, 55 pts. 85 pts completed neoadjuvant therapy according to the protocol, ***.90 pts underwent complete mastectomy, 4 had a breast-conservative surgery. After neoadjuvant treatment, investigator-assessed pCR rate was 27% (18%-36%, CI 95%). Interestingly, a significantly greater proportion of pCR rate was observed in the HR-negative group compared to HR-positive group (38% vs 13% respectively; p=0.007). Results from central review of pCR are currently in progress. Main Grade 3,4 toxicities were: neutropenia (61 pts),febrile neutropenia (36 pts),mucitis (23 pts) Neither gr3/4 HTA nor cardiac failure or proteinuria was reported. There were no treatment-related deaths. Conclusions: These data suggest that the use of BEV with neoadjuvant CT is active in HER2−negative IBC (pCR rate 27%), with an acceptable safety profile. Interestingly, the HR-negative subgroup is significantly more responsive than the HR-positive subgroup. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-20-01.

Background: Prognostic value of CTC detection in blood has been reported in primary breast cancer with a rate around 20% in the neoadjuvant setting. No correlation between CTC and pCR to neoadjuvant CT in operable breast cancer has been reported (Pierga CCR 2008, Riethdorf CCR 2010). Predictive value of CEC for response to anti-angiogenic agents is unclear. Methods: CTC and CEC were detected in 7.5 ml and 4 ml of blood respectively with CellSearch™ System in the neoadjuvant setting of HER2 negative IBC (T4d) patients (pts) enrolled in the phase II multicentre trial BEVERLY 1, evaluating bevacizumab (15mg/kg q3w) in combination with sequential neoadjuvant CT of 4 cycles of FEC followed by 4 cycles of Docetaxel. Patients received postoperatively 10 cycles of bevacizumab and hormonotherapy if tumor was ER+. Results: From 12/08 to 09/10, 101 pts were included, 96 were evaluable for pCR and 92 for CTC and CEC. Out of 96 pts, 51 (53%) had triple negative breast cancer (TNBC). At baseline, 37 pts out of 92 had ≥ 1 detectable CTC (40%, 95%CI 30–50%). At baseline, CTC level was not correlated with CEC level, neither with TNBC nor pCR. A drop in CTC incidence was observed from baseline to the 1st follow-up analysis after 4 cycles (p20/4ml) before C5 could be associated with a higher probability of pCR (Khi2 test, p=0.003). Conclusions: We observed a high CTC detection rate of 40% in HER2− IBC, including TNBC. CEC levels increased progressively during neoadjuvant treatment and decreased after its interruption. Baseline CTC and CEC levels were not predictive of pCR. Detection of CTC at 8 months of follow-up was associated with the absence of response to neoadjuvant chemotherapy. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-14-02.

Background: A 1228-probeset molecular classifier (EHT Dx14) able to discriminate benign breast lesions from breast adenocarcinoma with a 96% performance was previously identified in a panel of 71 fine-needle aspirate (FNA) samples of nodular breast lesions collected at Institut Gustave Roussy (IGR). The microarray which enabled the signature identification was based on the genomewide RNA analysis using the ExonHit's SpliceArray™ technology. The objective of the current study was to validate the performance of the EHT Dx14 molecular classifier on an independent cohort of FNA samples stored at the Centre de Ressources Biologiques at IGR. Patients and methods: The samples were collected from women undergoing investigations for suspicious nodular breast lesions classified BIRAD ACR 4 or 5 on mammogram and/or ultrasonography. The morphological cytological analysis of FNA samples classified the lesions as either benign, malignant or indeterminate. All samples were associated with a definite clinico-pathological diagnosis of the breast lesions. The first part of the study aims at confirming the sensitivity, specificity and overall performance of EHT Dx14 among lesions clearly classified by morphological cytology. The second part aims at establishing its performance among lesions with indeterminate cytological results. Results: In the first part of the study, the EHT Dx14 transcriptomic signature was applied on a set of 94 FNA consecutive samples categorized as benign (n=47) or adenocarcinoma (n=47) on cytology. The benign/malignant status of the lesions was confirmed by subsequent investigations or follow-up. Results of EHT DX-14 assessment of the samples were respectively 97.9% (95%CI: [88.7%-99.9%]) for sensitivity and 91.5% (95%CI: [79.6%-97.6%]) for specificity. Results on indeterminate FNA samples are being investigated in the second part of the study. Conclusion: This study did confirm the high performance of EHT Dx14 signature in identifying malignant from benign breast lesions in clinically suspect lesions. EHT Dx14 may provide adequate biological diagnosis in the absence of capability of morphological evaluation. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P1-07-10.

Background: CARMINA02 is a non-comparative multicenter phase II randomized trial evaluating the clinical response rate after up to 6 months of neoadjuvant endocrine therapy (NET) in HR+/HER2- patients with 1 mg anastrozole (Arm A) or 500 mg fulvestrant (Arm B). Secondary objectives included predictive markers of response and outcome. Between 2007 and 2011, 116 women with operable infiltrating breast adenocarcinoma T2-T4 N0-N3 M0 were randomized. Clinical response rates at 6 months (RECIST criteria) were 52.6% [95%CI 41-64%] in Arm A and 36.8% [95%CI 25-49%] in Arm B (Cancer 2016, in press). We aimed to identify the molecular predictive markers of resistance or sensitivity common to both treatments. Methods: Ninety tumor RNA from clinical responder (n=34) and resistant patients (n=23) treated in arms A or B have been sequenced with Illumina Hiseq2500 technology leading to 2x100-nt paired-end RNA-seq reads. These samples are from pre-treatment (29 in arm A, 28 in arm B) and post-treatment tumors (6 months after, 17 in arm A and 16 in arm B). Alignment was performed with Tophat_2.0.6. Differential gene expression was analyzed with the Differential Expression analysis for Sequence count data package. Gene fusion was detected with ChimeraScan, TophatFusion and DeFuse tools. Variant calling including variations, insertions and deletions was processed following GATK recommendations for RNAseq datas. Quantitative RT-PCR experiments were done to confirm RNA-seq expression results in patient samples not selected for RNA-seq analysis (validation cohort). Results: We first analyzed differentially expressed genes (DEGs) between responders and non-responders in pre-treatment or post-treatment samples to select potential predictive markers of response. We identified 51 DEGs before treatment common to anastrozole and fulvestrant. Among these 51 genes, SGK2 was the only gene more intensely expressed in responders than in non-responders. Then we compared DEGs between pre and post-treatment samples for responders or non-reponders for both treatment arms. SGK2 expression remains stable after treatment. Furthermore we identified 7 DEGs specific to responders and 11 DEGs specific to non-responders. Concerning genes fusion detection predicted by at least 2 tools, none was specific to a response type. We noted a higher number of fusions in non-responders samples. Variants detected by RNA-Seq are being confirmed by DNA-Seq using a home-made next-generation sequencing panel including 95 genes frequently mutated in breast cancers (analysis ongoing). Conclusion:High expression of SGK2, encoding a kinase induced in response to signals that activate PI3kinase, may represent a predictive marker of sensitivity to NET. DEGs associated with NET response or resistance belong to cell cycle, DNA replication and repair, cell death and drug metabolism. Ongoing DNA-seq datas will complete this genomic analysis. This research was conducted with support from AstraZeneca and Institut Curie. Citation Format: Callens C, Bessoltane N, Ngo C, Chemlali W, Becette V, Bernard V, Delattre O, Lemonnier J, Mouret-Reynier M-A, Andre F, Bieche I, Lerebours F. Genomic analysis to evaluate response to neoadjuvant anastrozole and fulvestrant in post-menopausal ER-positive HER2-negative breast cancer patients included in the UCBG CARMINA02 trial [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-04-09.

Background: High number of new drugs are targeting molecular alterations that occur in a few number of patients. Molecular screening that consists in the assessment of single molecular alteration during the screening phase of the therapeutic trial is not suitable. Indeed, this modality of screening is associated with a high rate of screen failure due to the low incidence of the molecular alteration. In order to optimize molecular screening, we have launched a clinical trial that aims at performing array CGH and hot spot mutations in patients with metastatic breast cancer. Patients and Methods. The trial plans to include 400 patients with metastatic breast cancer and included the first patient in may 2011. This trial is being sponsored by French federation of Cancer centers (UNICANCER) and involves 20 centers. Biopsy is being performed on metastatic site. Both frozen and FFPE samples are obtained. DNA extraction is being performed in the investigation center after control for the % of cancer cells. Array CGH and PIK3CA/AKT mutations (SANGE method) are being performed in four genomic platforms. A pilot study that included 106 patients has shown the feasibility of such technologies in the context of daily practice. Data from array CGH are being sent to a bioinformatician who forward the results on both pre-selected targets, DNA instability and some targets of interest selected based on log2(ratio) and function. Recommendations for trials are then being sent to each investigator. There is no limitation for previous lines and it is recommended to perform biopsy in patients who do not present progressive disease. This trial is being funded by French NCI (750 000 euros) Conclusion: SAFIR01 is a trial that aims at using high throughput technologies in order to drive patients with molecular alterations to specific therapeutic trials. Further amendments are being planned including implementation of high throughput sequencing and performance of functional testing. Further trial will compare this high throughput approach to standard methods for target identification. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr OT1-03-02.

Aim This trial was undertaken to assess the impact of late reopening of the infarct-related artery on left ventricular remodelling in post-myocardial infarction patients. Methods One hundred and fifty seven patients with recent myocardial infarction were routinely submitted to delayed (second week) catheterization. They also underwent systematic angioplasty of a significantly narrowed infarct-related artery with a suitable anatomy, or reopening of a totally occluded infarct-related artery, and repeat follow-up catheterization after 4 months. Changes in left ventricular ejection fraction, left ventricular volumes, and percent of regional hypokinesia were assessed over the study period. Results One hundred and thirty-two patients had two interpretable left ventriculograms and two interpretable coronary angiograms. At initial angiography, 56 out of 96 patients with a patent infarct-related artery were successfully submitted to percutaneous coronary angioplasty, of whom 25 had restenosis and eight had total reocclusion at follow-up angiography. Percutaneous transluminal coronary angioplasty was not attempted in the remaining 40 patients due to unsuitable anatomy in 18 or a nonsignificant lesion in 22. The infarct-related artery was totally occluded in 36 patients at initial angiography, and successfully reopened by means of angioplasty in 19, of whom seven showed a reocclusion at follow-up angiography. The independent predictors of left ventricular enlargement, identified by means of multivariate regression analysis, were initial stroke volume index 50 ml.m−2 (odds ratio=7·1, 95% confidence interval=[1·5; 25·8]), anterior infarct location (odds ratio=4·1, 95% confidence interval=[1·4; 11·5]) and reocclusion of the infarct-related artery (odds ratio=7·3, 95% confidence interval=[1·3; 27·3]). Angioplasty of a patent but significantly narrowed infarct-related artery was not found predictive. Conclusion This study demonstrates that reocclusion of a previously open infarct-related artery, as well as the initial low stroke volume index, enlarged end-systolic volume index and anterior infarct location are independent predictors of long-term left ventricular enlargement. These results emphasize the impact of long-term sustained patency of the infarct-related artery on the prevention of left ventricular dysfunction. The need for a larger randomized trial is recognised.

Background: The phosphatidylinositol 3-kinase (PI3K)/mammalian target of rapamycin (mTOR) pathway is often dysregulated in HR+ BC and is associated with resistance to endocrine therapy (ET). Alpelisib (BYL719; PI3Kα-specific inhibitor) and fulvestrant showed signs of antitumor activity in patients (pts) with estrogen receptor-positive (ER+), HER2– advanced BC (phase I), especially in PIK3CA-altered tumors (Janku et al. SABCS 2014, PD5-5). Methods: SOLAR-1 (NCT02437318) is a phase III, randomized, double-blind study in men and postmenopausal women with HR+, HER2– advanced BC. Pts are assigned to 1 of 2 cohorts based on PIK3CA tumor status (mutant vs non-mutant), and randomized 1:1 to oral alpelisib/placebo (300 mg once daily) and intramuscular fulvestrant (500 mg on Day 1 and 15 of Cycle 1; Day 1 of Cycles ≥2 [28-day cycles]) until disease progression or discontinuation. Randomization is stratified by presence of liver and/or lung metastases and prior CDK4/6 inhibitor therapy. Key inclusion criteria: recurrence or progression on or after AI therapy, ≥1 measurable lesion (RECIST v1.1) or predominantly lytic bone lesion, and ECOG performance status ≤1. Key exclusion criteria: symptomatic visceral disease or disease burden precluding ET, acute pancreatitis ≤1 year prior to screening or history of chronic pancreatitis, and prior therapy with fulvestrant, chemotherapy (except [neo]adjuvant), or PI3K/AKT/mTOR inhibitors. The primary and key secondary endpoints are progression-free survival (PFS; RECIST v1.1; local assessment) and overall survival (OS), respectively, in the PIK3CA-mutant cohort. Other secondary endpoints include PFS and OS in the PIK3CA non-mutant cohort, PFS (Blinded Independent Central Review; RECIST v1.1), the association between PFS and baseline PIK3CA status in circulating tumor DNA, overall response rate, clinical benefit rate, safety, and pharmacokinetics. The primary endpoint will be analyzed by a stratified log-rank test at one-sided 2% level of significance. Recruitment of the planned 560 pts is ongoing. Citation Format: Andre F, Kaufman B, Juric D, Ciruelos EM, Iwata H, Mayer IA, Rugo HS, Conte P, Liobl S, Rubovszky G, Inoue K, Tesch H, Lu Y-S, Ryvo L, Longin A-S, Mills D, Wilke C, Germa C, Campone M. SOLAR-1: A phase III study of alpelisib and fulvestrant in men and postmenopausal women with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2–) advanced breast cancer (BC) progressing on or after aromatase inhibitor (AI) therapy [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr OT2-01-04.

Objective: The purpose of this study was to determine whether differences in maternal serum cytotoxicity could be detected in different severities of gestational hypertension using a tumor necrosis factor sensitive cell line.Study Design: Serum was collected from 29 women with clinical signs or symptoms of gestational hypertension: 14 with severe preeclampsia (PET); 9 with pregnancy-induced hypertension (PIH); 6 with only elevated fibronectin and/or uric acid levels. Sera from 7 women at similar gestational age without any clinical signs of gestational hypertension were used as the normotensive controls. Maternal serum cytotoxic index [(1.0 – %FBS control) × 100] was assessed using the MTT assay in the L929 fibrosarcoma cell line. Differences in cytotoxic index and clinical parameters between groups were measured by ANOVA, and parameter relationships were examined by regression analysis.Results: Differences in cytotoxic index were detected between normotensive women (41.9 ± 6) and preeclamptics (65 ± 4.2)...

Background: Clinical remission of atopic dermatitis (AD) after intake of a low-salt water has recently been reported. Methods: We here investigated the effects of a low-salt water in AD patients with increased intestinal permeability to lactulose. From an initial group of 75 atopic patients, who undertook a treatment in a medical spa center, 12 were included because of abnormal urinary excretion of lactulose (>0.6%), which is absorbed in the enterocytic interspaces, at baseline. The excretion of mannitol with an extraenterocytic absorption was also studied as control. During 18 days, the patients followed a medical spa treatment course, including topical application and oral ingestion of a low-mineral water. Patients were asked not to modify their usual food regimen during the cure. At the end of treatment (day 18), the determination of the excretion of both sugars in urine was repeated. Results: At baseline, the mean urine excretion of lactulose in the study population was 1.21 ± 0.08 SEM. At the end of treatment (day 18), the lactulose concentration in urine diminished dramatically by 55% (mean 0.55 ± 0.09 SEM, p = 0.0005). Urine mannitol was also found to be reduced over the cure period, although to a lesser extent (30%, p = 0.04). Conclusion: This study indicates that low-mineral water normalizes the intestinal permeability of patients with AD.

Background: Although several studies have reported genomic characterisation of primary breast tumors, little is known about the genomic alterations of the metastatic tissues. Breast cancer patients who were prospectively enrolled in a trial (SAFIR01) underwent a biopsy of metastasis. The primary aim of this biopsy was to drive the treatment according to the results of whole genome CGH arrays and sanger sequencing on two genes (PIK3CA and AKT1). The results of the clinical trial were previously reported (André et al ASCO 2013). The secondary goal of the biopsies was to perform genomic characterisation of metastases in breast cancer patients. Here we report the analyses of such metastatic samples using next generation sequencing (NGS) approaches. Patients and methods: Two approaches were applied. In order to describe the incidence of targetable genomic alterations, we performed in depth targeted sequencing on 100 genes (200x coverage) using Illumina HiSeq 2000 (DNA vision). This analysis was performed on 240 metastatic samples. The second approach was more exploratory and aimed at discovering new genes involved in the metastatic process and/or resistance to therapies. In order to achieve this goal, we performed whole exome sequencing in 100 pairs of metastatic tissue (100x) and normal DNA (Integragen Inc, Hiseq platform). Finally, phosphor-S6K staining was performed in 300 samples in order to explore the activation status of mTOR pathway in metastatic disease, and to correlate with genomic data. Results: Targeted sequencing was performed on 240 metastatic samples in order to report the prevalence of targetable genomic alterations in metastatic breast cancer. Results are available for the first 159 samples. In addition to the already reported PIK3CA (26%) and AKT1 mutations (4%), NGS identified mutations of PTEN (4%), ERBB2 (2%), K-Ras (1%), ATM (2%), CDH1 (2%), GATA3 (2%), PTPN11 (1%), PTPRD (1%), ROS1 (1%). Results on the 81 samples, together with whole exome sequencing (n = 100) and phosphoproteins are being analysed and results will be available mid-november 2013. Conclusion: This is the first large study that aims at defining the genomic landscape of metastatic samples. Results will provide insights into the prevalence of targetable genomic alterations in metastatic tissue, together with candidate genes involved in the metastatic process and resistance to therapies. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr S6-07.

A 40-year-old man experiencing early satiety, a 25-1b weight loss, and intermittent nausea underwent an esophagogastroduodenoscopy, which demonstrated an ulcerated mass of the stomach extending from 43 cm to 53 cm from the incisors occupying the gastroesophageal junction, gastric cardia, fundus, and body. A biopsy of the gastric body showed diffuse infiltration of the lamina propria by poorly differentiated signet-ring cell adenocarcinoma. F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET) with computed tomography demonstrated minimally increased FDG uptake relative to normal tissues. Further study to determine the use of FDG PET in the evaluation of patients with gastric signet-ring cell carcinoma is warranted.

Background: Rare cancers sometimes present clonal evolution. Genomic characterisation of these rare entities could lead to identify new genes involved in the progression of more frequent cancers. In the present study, we have performed whole exome sequencing of three rare and aggressive histological subtypes in order to identify new targets that could subsequently be validated in common breast cancers. Methods: Whole exome sequencing was applied in ten pleiomorphic lobular carcinoma, seven micropapillary BC and eight metaplastic BC. Each cancer sample was matched with normal DNA from the same patient. Overall 50 samples were therefore sequenced. Tumor samples were selected to present >30% cancer cells. Histological diagnosis was confirmed by a BC pathologist. Whole exome sequencing (WES) was performed by BGI (Bejing Genomic Institute, Hong-Kong, China) using the Illumina Hiseq2000 platform (90bp paired-end reads, depth ≥80x). Genes recurrently mutated (with somatic mutations in >2 cases) were selected for validation. The validation by IonTorrent target sequencing was performed on additional formalin-fixed paraffin-embedded (FFPE) tumor samples (>30% of tumor cells). Results: Complete results of both WES and targeted resequencing validation are available for pleiomorphic lobular carcinoma. Nine of the 10 DNA pairs were eligible for WES. Genes already implicated in BC were significantly mutated: PIK3CA (3/9), TP53 (3/9) and CDH1 (3/9). Novel significantly mutated genes were identified: PYGM (2/9), EMR1 (2/9), ALDH1A2 (2/9), and DCLK1 (2/9). The validation using target gene sequencing was performed on the same 9 cases and on 19 additional confirmed pleiomorphic lobular BC cases (n = 28). For 8 out of the 9 cases with WES data, the results were reproducible on FFPE samples. Validation set confirmed recurrent mutations on PYGM genes (25%). PYGM is involved in glycogen metabolism. Further analyses on gene expression array revealed that PYGM is downregulated in most of the common BC samples, as compared to normal tissue, suggesting a role for this gene in the metabolism of common BC. WES data are available for micropapillary and metaplastic BC. HSPA8 (coding for a Heat shock protein) was found mutated in 3/7 of micropapillary samples. Validation on additional FFPE samples is ongoing and will be presented. WES of metaplastic BC revealed a high frequency of TP53 and PIK3CA mutations (50% in both cases). Conclusions: A novel mutated gene was identified in PLBC. PYGM is a gene involved in glycogen metabolism and was found downregulated in most of the common breast cancer. WES of micropapillary cancers suggests mutations on HSPA8, a gene encoding for a chaperon protein. Validation is ongoing. Deeper sequencing is ongoing for metaplastic BC that may help identifying new mutated genes in this multiclonal disease. Results will be presented. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr PD4-4.

Background:FGFR1 is a tyrosine kinase receptor involved in cell proliferation, migration and differentiation. FGFR1 gene is amplified in about 10% of breast cancer. Several studies have previously reported that FGFR1 inhibition could lead to a decrease in cell proliferation. Recent data obtained in prostate cancer have shown that FGFR1 activation could mediate epithelial to mesenchymal transition. In the present study, we evaluated the rational to target FGFR1 in clinical setting.Material and methods:We first assessed whether FGFR1 amplification was associated with a specific breast cancer phenotype. We assessed FGFR1 amplification by CGH array on 235 samples and by FISH (Z-2072, Spec FGFR1/Cen8 dual color probe, Zytovision®) on 260 samples, not included in the CGH part. We then assessed the functional effects of FGFR1 inhibitions in vitro. The antiproliferative effect of E3810 (FGFR1 inhibitor) was assessed both on FGFR1-non amplified (T47D/MCF7) and FGFR1 amplified (SUM44PE/MDA134) cell lines. Effect of FGFR1 inhibition was assessed on gene expression profiling.Results:When assessed by FISH, FGFR1 amplification was observed in 23/260 cases. Most of the samples (62%) exhibited an ER+/PR- phenotype, suggesting activation of intracellular kinase. FGFR1 amplification was observed in 24 out of 235 samples (10%). Interestingly, FGFR1 amplification was associated with gene gain in FGF3/FGF4/FGF19 (11q13) (55% versus 22%, p=0.05). FGFR1 inhibition by E-3810 inhibits MAPk phosphorylation, cell proliferation. A differential IC 50 for cell proliferation was observed between amplified cell lines and control cell lines. Preliminary data on gene expression profiling suggest that FGFR1 inhibition on breast cancer cells could lead to significant decrease in the Sox9 expression, a protein involved in epithelial to mesenchymal transition.Conclusion:FGFR1-amplified breast cancers present a discrete phenotype, including a high rate of ER+/PR- and high rate of FGF3/4/19 gains. Pharmacologic inhibition of FGFR1 leads to cell proliferation decrease and modulation of genes involved in EMT. These data suggest that there is a rational to target FGFR1 in clinical setting in patients with FGFR1-amplified breast cancer. Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 4170.

BackgroundBreast cancer is a collection of molecularly distinct neoplastic diseases and therefore, we hypothesized that p53 gene mutations may lead to different transcriptional changes in the different molecular subtypes and these may translate into subtype-dependent prognostic and predictive values.MethodsWe developed gene expression-based predictors of p53 status separately for estrogen receptor-positive (ER+) and -negative (ER-) breast cancers from a publicly available data set with known p53 mutation status (n=251). We validated the two signatures on an independent cohort of cancers (n=103) with known p53 functional status and tested their prognostic and predictive values on two other cohorts of breast cancers that received no systemic adjuvant therapy (n=255; n=198), and on one cohort of ER+ patients treated with adjuvant tamoxifen (n=277). We also examined if the p53 signatures were associated with chemotherapy sensitivity in ER+ and ER- cancers, respectively in two separate neoadjuvant data sets (n=233; n=103).ResultsWe developed a 39-gene p53 signature derived from 213 ER+ and a 30-gene p53 signature derived from 38 ER- breast cancers with no overlapping genes. External validation showed a sensitivity and specificity of 89% and 54%, respectively for the 39-gene signature in ER+ breast cancers; and 82% and 61%, respectively for the 30-gene signature in ER- cancers. The 39-gene signature was predictive of worse distant metastasis free survival (DMFS) in ER+ cancers with p53 dysfunction in both prognostic data sets (Hazard ratio (HR): 2.3 (95% confidence interval (CI):1.25-4.23, p=0.005 and HR:2.17 (95%CI:0.85-5.56, p=0.09). It remained predictive of worse DMFS even after tamoxifen adjuvant therapy (HR=2.43, 95%CI: 1.35-4.38, p Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 6122.

Abstract #2029 Background: Fine needle aspiration (FNA) for breast cancer diagnosis, a simple, quick and cheap procedure, is limited by a significant number of inconclusive exams. In the present study, we have analyzed whether the expression of a set of exonic probes could accurately diagnose breast cancer of FNA biopsies. Patients and methods: cDNA from 165 breast lesions (120 cancers), obtained by FNA in 2006-2007, and for which RIN score was > 6, was hybridized on splice arrayTM in two separate series (training/validation)(data set Andre et al SABCS 2008). We focused on 703 680 evidenced probe sets. The actual diagnosis of breast cancer was established by pathological examination at surgery. Associations between exonic probes and diagnosis were assessed by correlation coefficients (using t-tests) and a nearest centroid prediction rule was determined on the training set of 94 samples (70 cancers) in order to classify lesions as malignant or benign. Using the validation set of 71 samples (50 cancers), the performance of the prediction rule was evaluated and the proportion of explained variation (PEV) estimated in a logistic regression model (including only variables significant by a likelihood ratio test). Results: In a leave-one-out cross-validation analysis, the nearest prediction rule based on the expression of 1228 exonic probes (all probes with p Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 2029.

Abstract #6063 Background: p27 is a cyclin-dependent kinase (cdk) inhibitor that plays a role in cell cycle regulation. The expression of this protein, assessed by immunohistochemistry (IHC), has prognostic value for overall survival in breast cancer patients (Porter P.L., J Natl Cancer Inst., 2007). Based on this, we hypothesized that p27 could be a predictive factor for the efficacy of adjuvant anthracycline (A)-based chemotherapy in early breast cancer. Methods: Tumor samples from 823 patients, included at Institut Gustave Roussy (IGR) in two randomized trials comparing an A-based chemotherapy with no adjuvant treatment, were used to construct a tissue microarray. We previously reported predictive values of ER, Her2 expression and molecular subclassification using the same tissue array (Conforti R., Ann Oncol., 2007). Nuclear and cytoplasmic expression of p27 was assessed by the AQUA system. A p27 ratio of nuclear and cytoplasmic expression (p27 N/C ratio) was calculated for each tumor sample. The prognostic and predictive value of the continuous value of p27 N/C ratio was assessed by a Cox regression model adjusted for treatment, age, tumor grade, stage and ER expression (IHC) and stratified by trial. Results: p27 was assessable in 715 primary tumors. The p27 N/C ratio was significantly different according to the molecular subclasses (p=0.001) and was higher in Her2- versus Her2+ and in EGFR+ versus EGFR- tumors (p=0.0075 and p Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6063.