Search

Difference in pathologic complete response (pCR) rate after neoadjuvant chemotherapy does not capture the impact of treatment on downstaging of residual cancer in the experimental arm. We developed a method to compare the entire distribution of residual cancer burden (RCB) values between clinical trial arms to better quantify the differences in cytotoxic efficacy of treatments.The Treatment Efficacy Score (TES) reflects the area between the weighted cumulative distribution functions of RCB values from two trial arms. TES is based on a modified Kolmogorov-Smirnov test with added weight function to capture the importance of high RCB values and uses the area under the difference between two distribution functions as a statistical metric. The higher the TES the greater the shift to lower RCB values in the experimental arm. We developed TES from the durvalumab + olaparib arm (n = 72) and corresponding controls (n = 282) of the I-SPY2 trial. The 11 other experimental arms and control cohorts (n = 947) were used as validation sets to assess the performance of TES. We compared TES to Kolmogorov-Smirnov, Mann-Whitney, and Fisher's exact tests to identify trial arms with higher cytotoxic efficacy and assessed associations with trial arm level survival differences. Significance was assessed with a permutation test.In the validation set, TES identified arms with a higher pCR rate but was more accurate to identify regimens as less effective if treatment did not reduce the frequency of high RCB values, even if the pCR rate improved. The correlation between TES and survival was higher than the correlation between the pCR rate difference and survival.TES quantifies the difference between the entire distribution of pathologic responses observed in trial arms and could serve as a better early surrogate to predict trial arm level survival differences than pCR rate difference alone.

Background: We have previously shown that expression of the IL8/VEGFA-metagene eliminates the good prognostic effect of TILs in TNBC (PMID 21978456, 28750120). We also showed that the VEGFA metagene predicted response to neoadjuvant bevacizumab in the GeparQuinto trial (Karn 2017 SABCS #851166). The main cellular sources of the transcripts that comprise the VEGFA metagene are unknown since mRNA profiling of bulk biopsies contains signals from different cell types. Methods: Individual genes that comprise the VEGFA metagene were measured in bulk tissue- and single cell-RNA-Seq from breast cancer subtypes and normal cells on different platforms (Affymetrix n=4915, Agilent n=597, Illumina n=2433, RNA-Seq n=1215, Exome Capture RNA-Seq n=226, HTG-Seq n=243, sc-RNA-Seq n=24710). For blinded, orthogonal validation we performed immunohistochemistry. Effect of neoadjuvant chemotherapy with or without bevacizumab was studied by RNA-Seq and IHC on samples from GeparQuinto trial. SWOG S0800 (GSE114403), PROMIX (GSE87455), and GeparSixto trials were used for validation. TCGA was mined for mutations and somatic CNA. RNA-Seq from GeparNuevo was used for correlation with checkpoint inhibitor treatment. Results: We identified a stable core of six genes (VEGFA, ANGPTL4, ADM, NDRG1, DDIT4, CSTB) in different cohorts. Strong expression of this signature was mainly restricted to TNBC subtype and associated with poor prognosis within this subgroup. Single cell RNA-Seq of breast epithelial cells from 4 reduction mammoplasties and 4 TNBC revealed that these genes are coexpressed in individual epithelial cells and not associated with endothelial cells. In line with their presumed functions in cellular stress and hypoxia, immunohistochemistry revealed strong para-necrotic expression in TNBC. Moreover, high gene expression in TNBC was associated with mutations in DNA damage control pathways, somatic copy number alterations, and lower TILs. While chemotherapy led to downregulation, bevacizumab increased expression. In multivariate analysis, high pretreatment values predict pCR to both bevacizumab and chemotherapy (OR 2.40, P=0.006), which may be explained by sensitivity of tumors which are already under cellular stress. On the other hand, expression of the VEGFA metagene seems to create an immunosuppressive environment that counteracts the positive prognostic effect of TILs. In pre-treatment biopsies from the GeparNuevo checkpoint inhibitor trial we found a negative correlation of VEGFA metagene expression with the amount of the recently identified tissue-resident memory T cell subset (CD8TRM, PMID 29942092; P=0.002), while the subsequent increase of CD8TRM during treatment was larger in tumors with high VEGFA (P=0.019). Conclusions: Perinecrotic carcinoma cells under stress from hypoxia and or chromosomal instability are the source of the VEGFA metagene signature. Its predictive value in TNBC suggests estimating and reporting the amount of necrosis in the pathology report may be helpful in predicting response to preoperative chemotherapy, and could be used as stratification factor in clinical trials. The signature indicates an immunosuppressive environment and should be further studied in the context of immune therapies in combinations with anti-angiogenic treatment. Citation Format: Karn T, Denkert C, Sinn BV, Weber K, Nekljudova V, Rody A, Meissner T, Hatzis C, El-Balat A, Becker S, Solbach C, Untch M, Schneeweiss A, von Minckwitz G, Loibl S, Pusztai L, Holtrich U. Single-cell profiling identifies hypoxic carcinoma cells as source of an immunosuppressive VEGFA metagene [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 P3-10-01.

Introduction While immune infiltrates have already been extensively characterized in primary tumors (P), data on breast cancer metastases (M) remains limited. To this end we quantified and qualified the immune cells in a unique cohort of multiple matched P and M samples selected from an institutional breast cancer autopsy cohort. Patients and methods Twenty-three patients were selected from an institutional autopsy program (Semmelweis University, Budapest, Hungary) based on matched P and M sample availability (124 samples). All samples were centrally characterized for estrogen (ER), progesterone (PR) and HER2 receptors. The primary molecular subtypes were as follows: 9 ER+/PR+/HER2-, 8 triple negative and 6 HER2+. Ten patients relapsed ≤1 year after diagnosis and were further referred to as “early relapsers”, as opposed to the remaining qualified as “late relapsers”. Immunohistochemistry (IHC) was carried out against CD3/CD20 and CD4/CD8 in 21 patients (119 samples). Tumor infiltrating lymphocytes (TILs) were assessed on hematoxylin and eosin (H&E) and CD3-stained slides. Gene expression data were generated using the NanoString nCounter assay (PanCancer Immune Profiling Panel) for 11 patients (35 samples) and analyzedusing the R package NanoStringQCPro. The scores from published immune gene signatures were calculated as a weighted sum of the expressions of their genes. All samples were analyzed for 22 immune cell subtypes relative abundance using CIBERSORT. Results TILs assessed on H&E and CD3-stained slides were weakly correlated (Rho= 0.38, p Conclusion This is to the best of our knowledge the first study characterizing the immune infiltration in patients with multiple matched P and M samples. The results suggest that Ms have not only a globally lower immune infiltration as compared to Ps, but also a different immune composition. Additionally, Ms from late relapsers are more infiltrated as compared to early relapsers. The present data also uncovers not only important inter-patient but also intra-patient immune heterogeneity, which should be taken into consideration for optimal treatment decision. Citation Format: Szekely B, Bareche Y, Van den Eynden G, Salgado R, Buisseret L, Garaud S, Willard-Gallo K, Hatzis C, Szasz M, Kulka J, Larsimont D, Sotiriou C, Pusztai L, Desmedt C. Immune characterization of matched primary and multiple metastatic samples issued from an institutional autopsy cohort [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 PD5-10.

Background: In women with breast cancer receiving neoadjuvant chemotherapy, residual cancer burden (RCB) predicts distant recurrence and survival. In those with high risk tumors, locoregional recurrence (LRR) remains a concern, and has been associated with type of local therapy received. We evaluated the impact of local therapy on LRR in the ISPY-2 TRIAL. Methods: Data were analyzed in Stata 14.2, using Chi2 test, log rank test, and a Cox proportional hazards model. RCB was considered a categorical variable (0/1 versus 2/3), as described in prior publications. Breast surgery categories were lumpectomy +/- radiotherapy, or mastectomy +/- radiotherapy. Axillary surgery was defined as sentinel lymph node (SLN) surgery (≤6 nodes removed) or axillary dissection (>6 nodes). Results: Follow up data from the I-SPY2 TRIAL were available for 630 patients (median follow up 2.76 yrs, range 0.4-7.2). Type of local therapy was significantly associated with clinical stage at presentation, with stage III patients most frequently undergoing mastectomy + radiation (p In a Cox model adjusting for clinical stage, tumor subtype, surgical therapy, RCB status, nodal radiation, and age, significant predictors for LRR were tumor subtype and RCB status. Hazard ratio (HR) for LRR in those with RCB 0/1 was 0.39 compared to those with RCB 2/3 (95% CI 0.17-0.87, p=0.021). There was no difference in LRR between breast conservation and mastectomy; within the breast conservation group, those who had lumpectomy alone had higher hazard of LRR compared to those having lumpectomy + radiation (HR 3.1, 95% CI 1.1-9.2, p=0.043). Conclusions: Extent of surgical therapy was not associated with local tumor control, regardless of advanced tumor stage at presentation. Rather, tumor biology and response to therapy were the best predictors of LRR. These data highlight the opportunity to minimize the morbidity of extensive surgical therapy for patients with excellent response to systemic therapy. LRR rates by clinical features and treatment status FrequencyLRR RateP valueClinical Stage 0.5I240 (47.5%)5.8% II185 (36.6%)8.7% III80 (15.8%)6.3% Tumor Subtype 0.014ER+PR+Her2-161 (26.4%)3.1% ER+PR-Her2-56 (9.2%)3.6% Her2+176 (28.9%)6.3% Triple negative216 (35.5%)11.1% Local therapy 0.169Lumpectomy85 (13.5%)11.8% Lumpectomy with radiation198 (31.4%)5.6% Mastectomy173 (27.5%)5.2% Mastectomy with radiation174 (27.6%)8.6% Axillary surgery 0.23None5 (0.8%)20% SLN329 (52.2%)5.8% ALND296 (47%)8.5% Axillary radiation 0.535Yes42 (6.7%)9.5% No588 (93.3%)7.0% Axillary management 0.2No surgery or radiation5 (0.8%)20.0% SLN312 (50%)5.3% SLN+Axillary radiation17 (2.7%)8.3% ALND271 (43%)10.3% ALND+Axillary radiation25 (4%)5.4% RCB 0.0020/1293 (50.1%)3.8% 2/3292 (49.9%)10.3% Citation Format: Silverstein J, Suleiman L, Yau C, Price ER, Singhrao R, Yee D, DeMichele A, Isaacs C, Albain KS, Chien AJ, Forero-Torres A, Wallace AM, Pusztai L, Ellis ED, Elias AD, Lang JE, Lu J, Han HS, Clark AS, Korde L, Nanda R, Northfelt DW, Khan QJ, Viscusi RK, Euhus DM, Edmiston KK, Chui SY, Kemmer K, Wood WC, Park JW, Liu MC, Olopade O, Leyland-Jones B, Tripathy D, Moulder SL, Rugo HS, Schwab R, Lo S, Helsten T, Beckwith H, I-SPY 2 TRIAL Consortium, Berry DA, Asare SM, Esserman LJ, Boughey JC, Mukhtar RA. The impact of local therapy on locoregional recurrence in women with high risk breast cancer in the neoadjuvant I-SPY2 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 P2-14-01.

Background: Preclinical models suggest that in some HER2 overexpressed breast cancer cases, the extracellular domain of HER2, containing the binding region for trastuzumab (T) is not present due to proteolytic cleavage. Previous measurement of HER2 ECD and ICD in HeCOG 10/05 suggested that breast cancer cases that have low levels of ECD show decreased benefit from adjuvant T therapy (Carvajal et al, 2015). These cases may be more likely to show benefit from lapatinib(L), a small molecular tyrosine kinase inhibitor. NeoALTTO, an international multi-institutional trial comparing pre-surgical treatment with L vs T vs both, showed that pCR rates were around 25-30% for L or T but increased to 50% for both. This cohort allows us to test the hypothesis that patients with low ECD may benefit more from L than T. Methods: ECD and ICD were measured 382 cases using quantitative immunofluorescence (QIF) with domain specific antibodies (SP3, Spring biosciences for ECD and CB11, Biocare for ICD). Slides were stained and scanned, then measured using the AQUA method of QIF. All fields of view were scored and FOVs were averaged to generate a QIF score for each case. Index TMAs were used to standardize all autostainer runs and to define the QIF score cut-point that is equivalent to the clinical cut-point for the index TMA cases. Each case was then assigned to one of three groups; 1) High ECD and ICD, 2) Low ECD but High ICD, and 3) low ICD. Results: In the lapatinib arm, both group 1 and 2 showed a pCR rate of 25% (p>0.99), but in the trastuzumab arm, Group 1 pCR rate was 34% compared to 19% for group 2 (p=0.38). In the combination arm, group 1 showed a pCR rate of 59% compared to 29% for group 2 (p=0.046). In a logistic regression model for pCR, after adjustment for treatment arm and HR status the ECD/ICD status shows a significant predictive value (p = 0.002). Although this study is not powered for event free survival (EFS), at 6 years of follow up we found that patients in group 2 on the lapatinib arm show 92% EFS compared to 69% EFS in group 1 (p=0.17). The opposite trend is seen in both trastuzumab arms where EFS for group 1 on T showed 70% vs 62% for group 2 (p=0.57) and 77% vs 71% on the T+L arm (p=0.74). In a Cox regression model for EFS after adjustment for treatment arm and HR status, the ECD/ICD status is not significant (p=0.88) Conclusions: Cases with low levels of ECD compared to ICD appear benefit less than those with high levels of both ECD and ICD as assessed by pCR from treatments containing T. In contrast, those same patients also appear to benefit more, as assessed by EFS, on the L arm. These observations are consistent with the hypothesis that this assay could stratify benefit from L vs T based on the status of the ECD. Further assessment using this assay on the ALTTO cases is currently underway and may help confirm this observation. Citation Format: Rimm DL, Carvajal-Hausdorf D, Harbeck N, Fumagalli D, Rodrik-Outmezguine V, delaPena L, deAzambuja E, Huober J, Baselga J, Piccart M, Holmes E, Pusztai L. Low levels of HER2 extracellular domain (ECD) compared to intracelluar domain (ICD) in NeoALTTO may segregate benefit from lapatinib and trastuzumab in 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 P2-09-07.

BACKGROUND: As a neoadjuvant regimen for HER2-positive early breast cancer (BC), the use of two HER2-directed agents is more effective in producing pathological complete (pCR) responses than trastuzumab alone. Nevertheless, highly effective dual anti-HER2 combination may be unnecessary in patients who already benefit from a single agent. We previously reported that our 41-gene TRAR score is an accurate predictor of response to trastuzumab, with low scores being predictive of response to trastuzumab and favorable prognosis (Triulzi T. et al., 2015). PATIENTS AND METHODS: Tissue specimens from HER2-positive BC patients of Neo-ALTTO trial who received neoadjuvant trastuzumab and/or lapatinib plus paclitaxel were included in this study. Analysing RNA from fresh tissue using the 41-gene signature test, the area under the ROC curve (AUC) and the corresponding 95% confidence interval was computed to evaluate the predictive ability of TRAR score with respect to pCR, the primary endpoint of Neo-ALTTO. The prognostic role of TRAR score was investigated using a Cox regression model in univariate fashion. The patterns of Event Free Survival (EFS) according to the dichotomized TRAR score were estimated using the Kaplan–Meier method. RESULTS: The TRAR score was assessed for 226 of the 455 (49.7%) patients enrolled in the Neo-ALTTO study: 136 (60%) presented with T2 tumors, 188 (83%) with N0/1 and 128 (56.6%) with estrogen receptor negative BC. In details, basal TRAR score was available for 69, 79 and 78 patients assigned to neoadjuvant trastuzumab, lapatinib, and their combination, respectively. Overall, patients achieving a pCR showed significantly lower levels of TRAR score than those with residual disease (p CONCLUSION: Overall, we show that our 41-gene signature is accurate in predicting patient response to neoadjuvant HER2 targeted therapy in terms of pCR. In particular, low levels of TRAR score can identify a HER2-positive breast cancer subgroup highly responsive to trastuzumab as monotherapy for whom combination with other HER2-targeted drugs does not appear justified and may be one tool used for exploring de-escalating strategies without sacrificing outcomes. Citation Format: Di Cosimo S, Triulzi T, De Cecco L, Pizzamiglio S, de Azambuja E, Fumagalli D, Pusztai L, Harbeck N, Izquierdo M, de la Pena L, Huober J, Baselga J, Piccart M, Verderio P, Tagliabue E. Identifying clinically relevant subgroups of women with HER2-positive breast cancer: An analysis of Neo-ALTTO using the 41-gene TRAR score [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 P2-09-03.

Background: The tumor immune microenvironment is formed by many distinct and interacting cell populations, and its composition may predict patient's prognosis and response to therapies. Triple negative breast cancer (TNBC) is a heterogeneous disease in which different genomic subgroups have been described (Lehmann, et al. J Clin Invest.2011). Our aim was to integrate the composition of the tumor immune microenvironment with the molecular TNBC subtypes. Methods: We retrospectively analyzed the composition and the functional orientation of the immune microenvironment of 963 TNBC tumors clustered in 4 main TNBC subgroups (Basal-Like [BL1/2], Immunomodulatory [IM], Mesenchymal/Mesenchymal Stem-Like [MS], Androgen Receptor [AR]) from independent cohorts using transcriptomic profiling. TNBC were stratified based on 9 different immune signatures (Natural Killer [NK], Dendritic Cells [DC], T-Cells [TC], B-Cells [BC], T-Cytotoxic [TC], Interferon [IF], Nuclear Factor-kB [NF-kB], anti-tumor Macrophages [M1] and pro-tumor Macrophages [M2]. We validated our findings using immunohistochemistry. Results: We report that TNBC molecular subgroups and specific microenvironmental immune signatures are highly correlated. The IM was enriched of almost all the immune modules. On overall TNBC population, the TC (P = 0.01), IF (P = 0.007) and M1 (P = 0.001) signatures were associated with good prognosis. The BL subgroup with a good prognosis was characterized by overexpression of genes specific to M1-macrophages (P = 0.004) and TC-lymphocytes (P = 0.01). In contrast, the poor-prognosis MS expresses markers of cells of monocytic origin and significantly associated with the M2 signature (P = 0.01). The MS subgroup also displays an inflammatory and immunosuppressive signature. Pathological examination revealed that the MS subtype is characterized by a high density of pro-tumor macrophages that likely produce chemokines and cytokines which favor tumor-associated inflammation, resulting in a poor prognosis. In contrast, the AR exhibits low immune and inflammatory signatures. Conclusions: The distinct immune orientations of the TNBC molecular subtypes pave the way for tailored immunotherapies. Citation Format: Bottai G, Balazs G, Nagy A, Pusztai L, Szallasi Z, Reis-Filho JS, Santarpia L. Immune signatures define and affect prognosis in triple-negative breast cancer subtypes [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 P1-07-06.

Background:Anti-HER2 agents are associated with an increased risk of cardiotoxicity, occurring mostly during treatment and being generally reversible. To date, no cardiac biomarkers have been validated as predictive markers of cardiotoxicity in patients (pts) receiving anti-HER2 therapies. We evaluated high sensitivity (HS) troponin T (TnT) and N-terminal pro-brain natriuretic peptide (NT-proBNP) in pts treated with neoadjuvant anti-HER2 therapies to evaluate if early elevation of cardiac biomarkers predicts early or late cardiac events. Material and Methods: 455 pts were randomized to receive either lapatinib (154 pts), trastuzumab weekly (149 pts), or lapatinib with trastuzumab for a total of 6 weeks (152 pts). After this biological window, pts continued on the same targeted therapy plus 12-weekly paclitaxel until definitive surgery. After surgery, patients received adjuvant FEC (3x) followed by the same HER2 therapy as in the neoadjuvant phase for further 34 weeks (to complete 52 weeks in total). Eligible pts had an LVEF ≥50%. Cardiac biomarkers were centrally tested in plasma samples collected at baseline, week 2 (W2) and at surgery. Elevated biomarker was defined as either TNT >0.015 μg/L or NT-proBNP >125 pg/mL. The small number of pts and rarity of marker elevations allowed only descriptive analyses. Results: At a median follow-up of 3.8 years, 280 pts had cardiac biomarkers tested at baseline (61.5%) and NT-proBNP was elevated in 39 pts (13.9%) while TNT was elevated in only 1 pt (0.35%). From those 39 pts with NT-proBNP elevated at baseline, only 1 pt experienced a cardiac event. Also, hypertension was presented in 14 pts (24.5%) with elevated NT-proBNP. Serial measurements were available in 173 pts (61.7%) and NT-proBNP was elevated in 25, 10 and 21 pts at baseline, W2 and at surgery, respectively, while among 171 pts (61%) TNT was elevated in 1, 1 and 5 pts, respectively. There were 13 cardiac events in 11 pts in the entire trial. Five of these pts had a measurement of cardiac biomarkers at baseline (non-elevated NT-prBNP and TNT were observed in 4 and 5 pts), 9 of them had a measurement at week 2 (non-elevated NT-prBNP and TNT were observed in 8 and 9 pts) and 3 of them at surgery (non-elevated NT-prBNP and TNT were observed in 3 and 3 pts), respectively. The following table show the low specificity of the cardiac biomarkers. Patients without cardiac event (N)Patients with cardiac event (N)Patients with elevated markerElevated NT-proBNP Baseline38139Week 229130Surgery40040Elevated TNT Baseline101Week 2101Surgery606 During adjuvant phase, 13 pts experienced LVEF drop to Conclusions: Despite the small sample size, we demonstrated that cardiac biomarkers are rarely elevated in pts receiving anti-HER2 therapies, which is probably due to the absence of myocyte necrosis with respect to TNT, and the lack of chamber dilation in the case of NT-proBNP after exposure to these drugs. In addition, few cardiac events were present in this population not previously exposed to anthracycline-based chemotherapy; therefore the use of cardiac biomarkers in this population may be unnecessary. Citation Format: de Azambuja E, Bradbury I, Ewer M, Campbell C, Zardavas D, Ameels H, Baselga J, Huober J, Izquierdo M, Bozovic-Spasojevic I, Maetens M, Harbeck N, Pusztai L, Piccart M, Rodeheffer R, Sutter T. Cardiac biomarkers for early detection and prediction of trastuzumab and/or lapatinib-induced cardiotoxicity in patients with HER2 positive early breast cancer (BIG 1-06) [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 P4-21-09.

Background: A proportion of ER-positive, HER2-negative patients treated with both adjuvant endocrine and chemotherapy continue to remain high risk for recurrence. These individuals are the ideal patient population for future adjuvant trials. Methods: We used gene expression data from ER+/HER2- patients who received adjuvant endocrine and chemotherapy. We utilized a training cohort of multiple gene expression datasets (N=131; GSE16391, GSE21653, GSE17907, GSE19615, GSE16716, GSE45255) and a validation cohort (N=277; GSE25066) from similarly treated ER+ patients. All data sets were generated with Affymetrix HGU133 arrays. Cox proportional hazards regression was performed across all genes using all possible cutoff values between the lower and upper quartile. Genes were ranked according to the estimated hazard rates for relapse. The top three genes were combined into a final signature that used the mean expression of these genes as the predictive score. Statistical analyses were performed in the R environment. Results: The final signature capable to predict survival consisted of three genes, MX1, MBD4, and ZWINT. Each of these genes achieved high significance in a univariate analysis: MX1 HR=5.7 (p=1.8E-04) and HR=4.7 (p=1.3E-03), MBD4 HR=3.4 (p=2.8E-04) and HR=4.3 (p=4.4E-04), and ZWINT HR=4.4 (p=1E-03) and HR=3.7 (p=8.5E-03) in the training and validation cohorts, respectively. Using the mean expression of the three genes, the estimated hazard rates for survival were 20.5 (p=2.5E-05) and 17.5 (p=1E-04) in the two cohorts, respectively. Only one patient relapsed in each cohort. In a multivariate analysis including grade and lymph node status as covariates (size was not available), the three-gene signature retained significance (HR=13.2, p=0.011; and HR=13.9, p=0.01 in the two cohorts, respectively), while only lymph node status (HR=3.6, p=0.007) was significant in the validation cohort and no clinical variable was significant in the training cohort. Conclusions: High expression of a 3-gene signature can identify ER+ patients who remain at very high risk for recurrence despite current adjuvant endocrine and chemotherapy. Citation Format: Györffy B, Ocana A, Herman P, Hatzis C, Pandiella A, Pusztai L. Three-gene signature predictive of high residual risk of recurrence after adjuvant chemotherapy in ER-positive HER2-negative breast cancer [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 P6-09-49.

Background: Mutations or epigenetic silencing of BRCA1/2 genes result in DNA repair deficiency in a large proportion of TNBC cases. Yet it is unclear whether this deficiency is associated with increased chemosensitivity and improved benefit from standard-of-care chemotherapy. We systematically evaluated BRCA deficiency in TNBC using integrated DNA and RNA sequencing data and its association clinical outcome, exploring the potential role of tumor immune response. Patients and Methods: Whole-exome, DNA methylation, copy number variation, and RNA sequencing data from 102 stage I-III TNBCs were retrieved from The Cancer Genome Atlas (TCGA). Almost all patients received adjuvant taxane-anthracycline-cyclophosphamide (T-AC) chemotherapy and had >30 days of follow-up. The number of predicted neoantigens and an estimate of the level of immune cell activity for 77 of these tumors were previously published. Deleterious germline or somatic BRCA1/2 mutations were identified by majority voting on predictions of 5 variant scoring algorithms. Definition of BRCA1/2 deficiency (BRCA-D) included carrier of deleterious BRCA1/2 mutations or BRACA1/2 normal (BRCA-N) with wild type BRCA1/2 expression less than the maximum observed in mutation carriers. Normalized genomic mutation rate and mutant allele tumor heterogeneity (MATH) were computed as broad measures of genomic instability. Characteristics of BRCA-D vs N tumors were compared using the Wilcoxon rank test. Results: Twenty tumors (19.6%) had mutations in BRCA1, 6 (5.8%) in BRCA2, and 2 (1.9%) in both. Based on the expanded definition, 39 cases (38%) were characterized as BRCA1 deficient, 5 (4.9%) as BRCA2 deficient and 4 (3.9%) as deficient in both. BRCA-D tumors (47%) were associated with a significantly higher mutation rate (P=8x10-4) but had similar clonal heterogeneity (P=0.55) as BRCA-N tumors. BRCA-D tumors had excellent 4-year overall survival (100%) compared to 79.5% (95%CI: 66.6- 94.9) for BRCA-N tumors (log-rank P=0.02). BRCA-D tumors also presented a significantly higher number of predicted neoantigents (P=0.003), which resulted in increased level of immune cell activity. In contrast, low immunogenic TNBC tumors were underrepresented in BRCA-D (p=0.05) and showed potential signs of immunoediting (observed/expected number of predicted neoantigens < 1; p=0.07). Conclusions: Deleterious mutations in BRCA1/2 genes occur in 25% of TNBC tumors, but parallel quantification of wild-type BRCA1/2 expression identifies 47% of TNBC samples with double strand break DNA repair deficiency. These BRCA-D TNBC tumors are characterized by a significantly higher mutation rate and present a significantly greater number of neoantigens that result in increased immune cell activity. Our analysis suggested that enhanced immune activation could explain to a large extent the excellent clinical outcome in patients with BRCA-D tumors treated with standard-of-care T-AC chemotherapy. Citation Format: Jiang T, Safonov A, Bianchini G, Shi W, Wali VB, Pusztai L, Hatzis C. DNA repair deficiency enhances immune response and correlates with excellent clinical outcome in triple negative breast cancer. [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 P4-07-01.

Background: It is unclear if all breast cancer (BC) patients require baseline left ventricular function (LVEF) assessment prior to anthracycline based chemotherapy (ABC), and the approach is variable in clinical practice. Our objective is to determine the cost effectiveness of obtaining a baseline LVEF assessment prior to (neo) adjuvant ABC in clinical practice. Methods: We performed a retrospective analysis of the Yale Equilibrium Radionuclide Angiography (ERNA) database for 701 breast cancer patients who had a baseline ERNA scan prior to systemic therapy for an initial diagnosis of stages I-IV BC between July 2003 and May 2013. We found that 14 of 701 (2%) patients had a baseline LVEF < 50%. Age, pre-existing cardiac risk factors and coronary artery disease did not predict an abnormal baseline LVEF Results: Assuming that 20% of the unscreened patients with a LVEF < 50% will develop CHF if treated with ABC regimen without management of baseline cardiac dysfunction, the base case incremental cost effectiveness ratio (ICER) was determined to be 18,520 $USD/QALY. Sensitivity analysis suggested that the cost-effectiveness of baseline LVEF assessment is primarily driven by the prevalence of patients with LVEF < 50%, the incidence of CHF in this high-risk patient group if treated with ABC regimen, and time to CHF development. While our analysis did not reveal risk factors predictive of low baseline LVEF, our model's dependence on prevalence of LVEF < 50% demonstrates the importance of risk factor stratification. A hypothetical predictive marker which enriches the prevalence of an abnormal baseline LVEF 5-fold to 10% would result in a cost-effectiveness of 10,990 $USD/QALY. The model is less sensitive to the cost of baseline echocardiogram testing. Conclusion: Baseline LVEF assessment was found to be cost-effective under a willingness-to-pay threshold of $50,000/QALY. Our sensitivity analysis suggests that risk factor-guided LVEF baseline LVEF screening may increase the number of high-risk patients in the treatment population, thus further increasing the cost-effectiveness of baseline LVEF assessment. Citation Format: Safonov A, Hatzis C, Stratton J, Gross CP, Russell R, Pusztai L, Abu-Khalaf MM. A cost effectiveness analysis of baseline left ventricular function assessment for breast cancer patients undergoing anthracycline chemotherapy. [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 P6-11-03.

Introduction: Approximately 75% of stage I-III breast cancers are hormone receptor (HR) positive for which the standard of care is 5-10 years of adjuvant endocrine therapy, which has been shown to reduce recurrences and improve survival. Unfortunately, up to 40% of patients may not take the prescribed medication daily or may discontinue it early. Mobile health technology provides an opportunity to develop new innovative tools to identify women who are not taking medication as prescribed, to understand their barriers for adherence and to facilitate communication with providers to improve adherence. Methods: The objective of the BETA study was to develop a new bi-directional text messaging application that simultaneously assesses patient adherence to endocrine therapy and provides direct communication to the provider team. Our primary endpoint was to assess feasibility of the application and the secondary endpoints included adherence, side effects and their severity, and quality of life (QOL). The intervention consisted of 3 types of text messages to which patients responded: 1) daily, evaluating adherence, 2) weekly, evaluating medication-related side effects and their severity, and 3) monthly, evaluating barriers to taking the medication. After 3 months of participation, patients completed surveys assessing the tolerability and financial burden of the intervention and adherence to medication. Patients were eligible if they had stage I-III, HR-positive breast cancer, owned a cell phone, and were initiating endocrine therapy. Target enrollment is 100 patients. For comparison, 100 consecutive patients meeting the above criteria were identified retrospectively as historical controls; adherence was assessed via chart review. Results: Between November 2014 and May 2015, 62 patients (mean age 53.5 years) were enrolled and 25 had completed the study. Of those approached, 66% participated. Of those who completed the study, the application was found to be helpful by 63%; specifically, 76% felt the intervention was a reminder to take the medication, 96% felt it was easy to use, and 71% wanted to continue receiving text messages after the study ended. On average, patients spent 12 minutes with the application per week, 0% felt it took up too much time, and only 1 patient incurred text messaging fees. No patients withdrew from the study and only 1 patient did not adhere to treatment (as defined by ≥ 80% adherence). None of the enrolled patients discontinued endocrine therapy, compared to 9% of historical controls. Side effects were common: hot flashes/night sweats (61% of patients), joint aches/pains (56%), and vaginal symptoms (29%) were reported. Severe side effects (reported by 29% of patients) prompted a return phone call to the patient. The study is ongoing and final results will be available by December 2015. Conclusion: We developed a new bi-directional text messaging intervention to assess adherence to endocrine therapy that provides real-time feedback to providers. Patients found the application helpful, easy to use, and not time consuming. Our tool is scalable for large population-based trials. Citation Format: Epstein LN, Jhaveri AP, Han G, Abu-Khalaf MM, Hofstatter EW, Sanft TB, DiGiovanna MP, Silber AL, Adelson KB, Chung GG, Pusztai L, Gross CP, Mougalian SS. Development of an interactive text messaging tool to improve adherence with adjuvant endocrine therapy: Breast cancer endocrine therapy adherence (BETA) pilot 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 P5-11-03.

Background: The presence of tumor-infiltrating lymphocytes (TILs) is a prognostic factor in triple-negative breast cancer (TNBC). In particular, cytotoxic CD8+ cells are associated with favorable outcome in TNBC. Recently, the integrin CD103 have been suggested as a potential prognostic marker in a small cohort of triple-negative/basal-like breast cancer (Wang et al, Clin Cancer Res 2016). In this study, we aimed to evaluate the functional status and clinical relevance of CD8+ and CD103+ TILs in TNBC. Methods: A large collection of formalin-fixed, paraffin-embedded tissues was retrospectively collected from 210 patients with primary invasive ductal TNBC. The presence of CD8, CD103 and the checkpoint receptor PD-1 was assessed by immunohistochemistry. Whole tumor slides were independently assessed by two pathologists blinded for patient characteristics and outcome. Cases where ≥5% of TILs expressed CD103 or PD-1 were considered positive. Statistical analyses were performed using Spearman's correlation, Kaplan-Meier and Cox regression analyses. Results: We found that CD103+ cells mostly co-expressed the CD8 marker, and were preferentially distributed within tumor epithelium. No consistent correlation was found between the presence of CD103+ lymphocytes and the expression of its ligand E-cadherin by TNBC cells (rs=0.365). CD8+ lymphocytes were consistently associated with better relapse-free survival (RFS) and overall survival (OS) in both univariate (HR=0.61; P=1.06E-02 for RFS; HR=0.59; P=1.93E-02 for OS) and multivariate analysis (HR=0.65; P=2.45E-02 for RFS; HR=0.63; P=3.96E-02 for OS). The presence of CD103+ lymphocytes significantly correlated with prolonged RFS and OS in univariate analysis only (HR=0.82; P=4.12E-02 for RFS; HR=0.85; P=4.93E-02 for OS), and a trend for longer RFS was also observed in univariate analysis for PD-1 (HR=0.71; P=5.22E-02). Interestingly, a subset of TNBC showed co-expression of CD103+/PD-1+ in lymphocytes localized to the intraepithelial areas of the tumor. Conclusions: The expression of CD103 on CD8+ T cells in direct contact with cancer cells may identify a subpopulation of cells with potent cytolytic activity. The interaction between CD103+ cytotoxic lymphocytes and cancer cells is mediated by mechanisms other than the binding to E-cadherin. Even though CD103 has a role in mediating an effective anti-tumor immune response, its presence alone may not be sufficient to impact the outcome of TNBC patients. Immunomodulatory therapies may be useful to boost the anti-tumor activity of potentially quiescent CD103+/PD-1+ cells in a subgroup of TNBC. Citation Format: Bottai G, Loi S, Sotiriou C, Roncalli M, Pusztai L, Reis-Filho JS, Santarpia L. CD103+/PD-1+ T-cells identify a subset of triple-negative breast cancer eligible for targeted checkpoint inhibition [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 P2-04-16.

Background The prognostic and predictive relevance of different components of the adaptive immune system in HER2-positive, TN and high proliferation ER+/HER2- subtypes have been shown. In this scenario, the remarkable therapeutic efficacy of new immunomodulatory drugs targeting the immune checkpoints (i.e. CTLA4) provides attractive opportunities. We aimed to investigate the prognostic value of genes related to CD28 and CTLA-4 signaling pathways in early breast cancer patients. Methods We assessed the affymetrix gene expression data of 789 untreated, node-negative, early breast cancer patients. Expression of CTLA4, CD28 and the two corresponding natural ligands CD80 (B7-1) and CD86 (B7-2) were evaluated. The sum of these two ligands was combined as a unique marker (CDligs). The prognostic value of the three biomarkers and their interaction were assessed independently in each molecular subtype (ER-/HER2-, HER2+ and high proliferation ER+/HER2- [ER+/HER2- HP]). Multivariate analysis was adjusted for PGR, proliferation and other immune related genes (Tcell, IGG, MHC II, MHC I, Interferon and STAT1) (Gianni L, SABCS 2012). Distant event free survival was the outcome of interest. Results A significant moderately positive correlation was observed between each biomarker. Expression of CTLA4 and CDligs was higher in ER-/HER2- and HER2+ compared to luminal subtypes. High expression of CTLA4 and CD28 were associated with lower risk of relapse in HER2+ (p = 0.008 and p = 0.03, respectively) and ER+/HER2-HP (p = 0.03 and p = 0.008, respectively). High expression of CDligs was also associated with lower risk of distant event in HER2+ group only (p = 0.01). No interactions were observed between CD28 and CTLA4 in any subtype. In both TN and ER+/HER2- HP subtypes, a significant interaction between CDligs and CTLA4 (interaction term p = 0.009 and p = 0.025, respectively), CDligs and CD28 (interaction term p = 0.028 and p = 0.008, respectively) and CDligs and Tcell metagene (interaction term p = 0.006 and p = 0.019, respectively) was observed. For instance, in high-CTLA4 group (above median cut point), higher expression of CDligs was associated with significantly lower risk of distant relapse in TN [HR 0.41 (0.21-0.82); p = 0.012], HER2+ [HR 0.25 (0.08-0.72); p = 0.011) and ER+/HER2- HP [HR 0.40 (0.29-0.85); p = 0.01] subtypes. Instead, in the low-CTLA4 group, CDligs was not associated with distant relapse in any subtype [TN, HR 1.80 (0.28-3.96) p = 0.141; HER2+, HR 0.69 (0.24-1.99) p = 0.495; ER+/HER2- HP, HR 1.28 (0.70-2.35) p = 0.425]. These interactions retain statistical significance in multivariate analysis including PGR, proliferation scores and other immune metagenes. Conclusions Genes related to the CD28/CTLA4 immune checkpoint are associated with risk of distant relapse independently of other biomarkers and immune related metagenes. A significant interaction between expression of ligands (CD80/CD86) and the two corresponding receptors CD28 and CTLA4 was found in TN and ER+/HER2- HP subtypes. These data confirm the complex interactions involved in the adaptive immune system regulation which may not be fully appreciated without consideration of biomarker interactions. They also provide additional evidence for exploring the use of immunomodulatory drugs in breast cancer. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P2-11-04.

Aim Both hypothermia and central nervous system (CNS) drugs may alter the predictive accuracy of amplitude-integrated electroencephalography (aEEG) in hypoxic–ischaemic encephalopathy (HIE). The aim was to assess the predictive value of aEEG in hypothermia-treated HIE infants. Furthermore, we intended to investigate the association of cumulative doses of CNS drugs with aEEG recovery. Methods Seventy term HIE infants treated with hypothermia for 72 h were continuously monitored by single-channel aEEG. Doses of administered morphine, phenobarbitone and midazolam were recorded. Poor outcome was defined as death or severe neurodevelopmental delay at 18-24 months (Bayley Scales of Infant Development II), good outcome as absence of these criteria. Results Poor outcome n = 26, good outcome n = 44. Positive predictive values (PPV) of an abnormal background pattern to predict poor outcome were 0.5 at 6 h; 0.65 at 24 h; 0.82 at 48 h and 0.92 at 60 h. All infants who developed sleep–wake cycling (SWC) had a favourable outcome; the nondevelopment of SWC resulted in a PPV of 0.73 for a poor outcome. Cumulative doses of the investigated drugs did not differ between infants having an onset of a recovered background pattern before or after 24 h. Conclusion Amplitude-integrated electroencephalography provides reliable prediction of outcome from the 48th hour during hypothermia in HIE infants. Commonly used CNS drugs in HIE infants do not significantly delay aEEG recovery.

Objectives: Different clinical behaviors are observed in tamoxifen-treated and untreated ER-positive, HER2-negative highly-proliferative breast cancer (BC) that demonstrate either high (highERS) or low (lowERS) expression of estrogen-related genes (Bianchini SABCS 2011). LowERS tumors are intrinsically endocrine resistant and at significant risk of relapse in the first 5 yrs after diagnosis. We studied lowERS and highERS BC in pts treated with neoadjuvant chemotherapy (NAC) and examined prognostic and predictive markers in the highest risk group of lowERS BC. Methods: We examined affymetrix gene expression data from 193 ER+/HER2−, high proliferation BC from pts treated with taxane-anthracycline-based NAC followed by endrocrine therapy. Previously defined cut-offs for markers of proliferation (MKS), and estrogen-related genes were applied (Bianchini SABCS 2011). Within the lowERS group, we examined pts treated with no systemic adjuvant therapy (n = 137; 50 events); adjuvant tamoxifen-only (n = 141; 36 events); and NAC (n = 127, 27 RCB0/I). We performed gene enrichment analysis for 2617 gene sets with known biological function (by 5000 random permutations). Primary endpoints were distant event free survival (DEFS) with follow-up censored at 5-yrs and pathological response (pathR) using the residual cancer burden (RCB) (Symmans JCO 2007). Results: The median follow-up of the NAC series was 3.1yrs. The DEFS at 4yrs was 0.94 [0.87–1.00] and 0.70 [0.60–0.81] in the high and low ERS groups, respectively (p = 0.004) (despite the higher rate of pathR (RCB0/I) to NAC in the low ERS group (9.5% and 21.9%; p = 0.04)). The pathR was prognostic in the lowERS group [HR 9.1 (CI 1.23–67.4); p = 0.009] but not in highERS (p = 0.485). In contrast, a different outcome was observed in BC with RCBII-III, were the 4-yrs DEFS was 0.93 [0.86–1.00] and 0.61 [0.49–0.76] in high and low ERS group, respectively (p = 0.0007). In the lowERS group there was substantial overlap in biological functions associated with prognosis in both tamoxifen-treated and untreated pts. At a conservative threshold of p < 0.0005, 38 gene sets were significant (31 good-prognosis involved in adaptive immune function, inflammation and chemotaxis and 7 poor prognosis gene sets involved in regulation of nuclear division and cell polarity). Only proliferation-related gene sets were predictive of RCB0/I. MKS as a continuous variable was predictive of pathR beyond clinical variables [OR 5.43 (2.04–16.1); p = 0.001]. Conclusions: Among ER+/HER2−, high proliferation BC the highERS group showed a low pathR rate with excellent prognosis even if there was residual disease. The lowERS group showed a higher rate of pathR which was significantly prognostic for good outcome. Lack of pathR in this group predicted for very poor prognosis despite subsequent adjuvant endocrine treatment. The clinical behaviour and aggressiveness of this subgroup is similar to triple negative tumors. The prognostic relevance of immune function provides a rationale for testing immunotherapeutic strategies in this subgroup. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-10-10.

Background and Aims: MicroRNAs (miRNAs) are short non-coding RNA whose expression levels have diagnostic and prognostic implications in cancer. However, the relevance of aberrantly expressed miRNAs to breast tumour-immune response has not been established. We aimed to assess the association between deregulated miRNAs and tumour immune response in breast molecular cancer subtypes. Methods: One hundred and seventy ductal infiltrating breast carcinomas (BC) (ER+/HER2− n=85, HER2+ n=40, ER−/HER2− n=45) were evaluated for the presence of cellular immune components by IHC assay. CD3, CD20 were used as markers for T lymphocytes infiltration (TIL), and B lymphocytes infiltration (BIL), respectively, whereas CD68 for the presence of tumour-associated macrophage (TAM). Total RNA (including miRNA) was extracted from all tumours and 32 highly specific miRNAs and 276 immune genes were evaluated by array and qRT-PCR assays. We derived an integrated immune miRNA-regulatory gene profile to tease out relationships of these genes as potential targets of specific miRNAs. In vitro experiments on BC cell lines were performed to confirm potential links between deregulated miRNAs and immune genes. Results: Lymphocytes infiltration (LI) was most common in HER2+ and ER−/HER2− BC samples (P < .004 and < .001) although B and T cells components were differently distributed in these 2 BC groups. TAM was a frequent event in ER−/HER2− BC subtype (P < .005). By gene and integration analyses we identified important miRNA immune-modulated target genes associated with prognostic factors and with molecular BC subtypes. We focused our studies on miR-181 and miR-135a that were significantly associated with ER−/HER2− (P < .001) and HER2+ BC (P < .002), respectively. BCs with abnormal expression of miR-181 and miR-135a showed an important association with distinct TIL and TAM components and specific immune genes. Importantly, preclinical data on BC cell lines demonstrated that miR-181 and miR-135a were able to regulate important cell immune-modulators such as SEMA3B and STAT6 signalling pathways, respectively. These 2 cellular pathways in turn were capable to regulate specific cascades of immune gene networks. Conclusion: We identified specific miRNAs associated with immune cellular components and immune-related tumour response in the 3 main molecular BC subtypes (ER+/HER2−, HER2+ and ER−/HER2). Aberrant expression of miR-181 and miR-135a can contribute to BC tumour immune response programs by regulating important gene targets such as SEMA3B and STAT6 signalling pathways in ER−/HER2− and HER2+ tumours, respectively. Modulation of miRNAs activating tumour immune response programs may have important implications for novel therapeutic strategies in BC. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P5-10-01.

Background Bone metastasis (BM) is one of the most common sites of distant metastasis for breast cancer. We hypothesized that its occurrence can be predicted if a clinical nomogram can be developed, thus allowing for selection of patients at high risk for BM. Patients and Methods Medical records of patients with non metastatic breast cancer were prospectively collected for the period between January 1997 and February 2007 at the M.D. Anderson Cancer Center (Texas, USA). A multivariate logistic regression analysis of selected prognostic features was done. Based on Cox proportional hazards regression model, a nomogram to predict BM was constructed and validated in an independent cohort of 579 patients with non metastatic breast cancer at time of diagnosis treated at the Tenon Hospital (Paris, France) between January 2003 and December 2005. Results: Of 4175 patients with non metastatic breast cancer, 314 developed subsequent BM. Age, T stage, lymph node status, lymphovascular space invasion, breast cancer molecular subtype, adjuvant hormonotherapy were significantly and independently associated with subsequent BM. The nomogram had a concordance index of 0.69 (95% CI, 0.68 to 0.70) in the training set. The validation set showed a good discrimination with a concordance index of 0.65 (95% CI, 0.57 to 0.72). At 3, 5 and 7 years, the nomogram was well calibrated. Conclusion: We have developed a robust tool that is able to predict subsequent BM in patients with non metastatic breast cancer. Selection of an enriched patient population at high risk for BM will allow to practice individualized therapeutic strategies, an adapted medical supervision and will facilitate the design of trials aiming at its prevention with the use of biphosphonate treatment. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P2-12-06.

Background Used upfront or after 2 to 3 years (yrs) of tamoxifen (TAM) or as extended treatment after 5 yrs of TAM, aromatase inhibitors (AIs) are associated with less recurrences than with 5 yrs of TAM. However, up front AIs were not superior to TAM→AI sequence in node negative (N-) tumors (TEAM study, SABCS 2009; BIG1-98 trial, NEJM 2009). The observation is consistent with the hypothesis that early relapses may be more frequently due to intrinsic endocrine resistance irrespective of the use of TAM or AIs whereas late relapses depend on acquired resistance. Methods Affymetrix HGU133A-based gene expression profiles from two adjuvant datasets (n=556) of patients treated with 5 yrs TAM were evaluated and included 285 N-, 247 N+ and 24 Nx estrogen receptor-positive cases. A proliferation score based on the expression of 12 mitotic kinases (MKS) (Bianchini, Cancer Res 2010) and a 4-gene estrogen-related score (ERS) adopted from the Oncotype DX Recurrence Score were assessed. Median cut-off points were used for both biomarkers. Pattern of relapse according to marker status were evaluated in three distinct time cohorts (0–2.5 yrs, 2.5–5 yrs, 5–10 yrs). Outcome was assessed according to distant relapse rates. Results A violation of the proportional hazards assumption was found for both scores indicating time-dependent effects. The table shows distant relapse rates according to biomarker group within each time cohort. Among low proliferation tumors (lowMKS), the highERS group had low and steadily increasing recurrence rates, while the lowERS group showed a higher risk of relapse that increased continuously over the time. Among the highly proliferative tumors (highMKS), the high endocrine-sensitive group (highERS) had a lower risk of relapse within the 0–2.5 yrs compared to lowERS cancers. The difference is not persisting at the longer follow up of 2.5–5 yrs, and at 5–10 yrs the highERS group has higher risk of relapse. The pattern of relapse was similar for N- and N+ cancers. However, in the 0–2.5 yrs interval all recurrences occurred in the highMKS/lowERS among N- cancers, while only 57% of recurrences did occur in N+ cases in the same biomarkers group (p=0.012). Discussion Highly proliferative, high endocrine-sensitive (highERS) cancers are at the greatest risk of late relapse and extended endocrine therapy beyond 5 yrs may be useful. Low proliferation, low ERS cancers also remain at substantial risk for late recurrences. Between 0 to 2.5 yrs all recurrences in N- tumors and close to 60% in N+ were in high proliferative/low endocrine-sensitive tumors. This difference could explain the beneficial trend for up-front AIs vs. sequential TAM→AIs in the N+, but not in N- subgroups of the TEAM and BIG1-98 trials. It also is consistent with the hypothesis that the highly proliferative low ERS group includes tumors intrinsically resistant to both AIs and TAM. Risk stratification by these two simple metrics may improve the readout of future adjuvant clinical trials with endocrine drugs, and may influence their design. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr S1-7.

Background: Bone metastasis-specific genes in breast cancer have been reported without considering the significant differences in ER status between bone and non-bone metastases. The aims of this study were to validate genes that had been reported as bone metastasis-specific genes using our data set and to identify bone metastasis-specific genes on the basis of biological differences between ERpositive and ER-negative breast cancers. Patients and Methods: We used Affymetrix GeneChip arrays to analyze tumor samples obtained from 365 primary invasive breast cancer patients who underwent surgery from 1999 to 2008. We excluded patients with HER2−positive breast cancer (normalized HER2 mRNA expression [probe set 216836_s_at] > 12.54). We classified the samples into 3 cohorts according to first metastatic site: bone, non-bone, or no metastasis. Differential expression of genes between bone and non-bone cohorts that were differentially expressed was identified using the Cox proportional hazards model, and gene sets was assessed using gene-set analysis. Results: Of the 365 patients, 34 (9.3%) were included in the bone cohort and 32 (8.8%) in the non-bone cohort. Two hundred fourteen (58.6%) had ER-positive and 151 (41.4%) had ER-negative breast cancer. First, we performed gene-set analysis using 5 gene sets that had been reported to be associated with bone metastasis. One gene set, which had been detected using an ER-negative breast cancer cell line, was validated as predicting bone metastasis in ER-positive breast cancer. None of the 5 gene sets predicted bone metastasis in ER-negative breast cancer. We then determined the levels of individual genes associated with bone metastasis by ER status using all 16,712 probe sets filtered by average gene expression level. When we analyzed all patients without any stratification by ER status, as in previous studies, 592 probe sets were significantly overexpressed in the bone cohort compared with the non-bone cohort, with a false discovery rate of ≤0.05. However, when we analyzed ER-positive and ER-negative breast cancers separately, no genes were found with significant differences between bone and non-bone cohorts. Finally, we used 2,246 functionally annotated gene sets assembled from Gene Ontology to examine possible biological differences between bone and non-bone cohorts. In the bone cohort, 151 and 125 gene sets were significantly overexpressed in ER-positive and ER-negative breast cancers, respectively (P ≤ 0.05). Ppathways related to Cellular growth and proliferation, and intracellular and second-messenger signaling were overexpressed in ER-positive breast cancer, whereas pathways related to nuclear receptor and cytokine signaling were overexpressed in ER-negative breast cancer. Most bone-metastasis-related pathways were different in ER-positive and ER-negative breast cancers (91.4% and 89.6% of the gene sets, respectively). Discussion: No genes were found that can predict bone metastasis. ER-positive and ER-negative breast cancers have different biological potentials for bone metastasis. Therefore, we need to assess the prediction model of bone metastasis based on the biological features for each ER status separately. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-16-02.

Triple-negative breast cancers (TNBC) lack the expression of estrogen, progesterone, and the human epidermal growth factor 2 receptors and have limited treatment options. We hypothesized that genes that are frequently amplified or overexpressed in these cancers are functionally important for the growth and survival of triple-negative breast cancer. Materials/Methods: We identified genes overexpressed in TNBC in 3 different human breast cancer gene expression data sets (n=294, n=286, n=198). We assessed the functional importance of the consistently overexpressed genes using siRNA screen on 18 breast cancer cell lines (10 ER-, 8 ER+) in vitro. We targeted each gene with 4 different siRNA constructs separately and each experiment was performed in triplicate. The genes with the greatest TNBC-specific inhibitory effect after siRNA down regulation were selected for further mechanistic and signal transduction studies. Results: 684 genes showed consistent and highly significant overexpression in TNBC compared to receptor-positive cancers in all 3 data sets. siRNA suppression of 161 of these genes inhibited cell growth significantly more in the ER-negative compared to ER-positive cells by at least 1 of the 4 siRNAs, 27 genes showed similar effect with 2 or more siRNAs and for 2 genes 3 of the 4 siRNAs showed preferential growth inhibition in ER-negative cells. These two genes were VLDLR (very low-density lipoprotein receptor) and LRP8 (low density lipoprotein receptor-related protein 8). We validated the siRNA screen results and confirmed down regulation of the mRNA and protein levels for VLDLR and LRP8 in 4 different ER-negative cell lines and showed that siRNA inhibition can be rescued by co-transfection of the receptor genes. Reanalysis of gene expression data also indicated expression of the LRP8 and VLDLR ligands, Reelin or ApoE, in both breast cancer tissues and in cell lines. We next demonstrated that exposure to Reelin and ApoE stimulates the growth of ER-negative cells in vitro. The stimulatory effect of ApoE was isoform dependent ApoE2 (Cys112/Cys158;) has the lowest receptor binding affinity and showed no growth stimulation, ApoE3 (Cys112/ Arg158) had modest 50–60% growth stimulation and ApoE4 (Arg112/ Arg158) had the greatest stimulatory effect (300-400%). Suppression of the expression of either LRP8 or VLDLR or exposure to RAP (an inhibitor of ligand binding to LRP8 and VLDLR) abolished this ligand-induced proliferation. ApoE4 stimulation results in the transcriptional upregulation of genes involved in proliferation, metabolism, and inflammatory signaling pathways. ApoE4 stimulation increases expression of proteins involved in MAPK/ERK pathway, DNA damage repair, and inflammation. Conclusions: We show that theApoE - LRP8/VLDLR ligand receptor system is overexpressed in human TNBC. We also demonstrated that this receptor system mediates a strong growth promoting and survival function in TNBC cells in vitro. Interestingly, allelic imbalance favoring ApoE4 expression (E3/E4 or E4/E4) has been linked to higher risk of developing early onset breast cancer, which is primarily TNBC. We propose that inhibitors of LRP8/VLDLR signaling may be clinically useful therapeutic or preventive agents for TNBC. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-17-01.

This abstract was withdrawn by the authors. Citation Format: Di Cosimo S, Appierto V, Pizzamiglio S, Baselga J, Piccart M, Huober J, Izquierdo M, de la Pena L, Hilbers F, de Azambuja E, Untch M, Pusztai L, Pritchard KI, Nuciforo P, Salomon AV, Symmans FW, Apolone G, de Braud F, Verderio P, Daidone MG. Withdrawn [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 P4-01-03.

Background: Several gene signature-based predictors of response to targeted drugs have been proposed in the literature but none has been prospectively tested as patient selection tools in the clinic. The goal of this trial is to assess the positive predictive value of 3 conceptually different multi-gene signatures as predictors of response to the multitargeted kinase inhibitor dasatinib. Methods: This clinical trial requires biopsy of a metastatic lesion for gene expression profiling and employs a parallel, multi-arm, two-step, phase II design. Three markers are assessed including a (i) cell-line derived dasatinib-sensitivity signature, (ii) a src-pathway activity signature and (iii) a dasatinib target index calculated as the weighted average expression of all known dasatinib targets. Only markerpositive patients are treated with dasatinib 100 mg po daily and each marker arm is considered as a separate study with early stopping rules for futility (minimum sample size 9, maximum sample size 40/marker arm). A predictor is considered worthy of further study if the clinical benefit rate (i.e. positive predictive value) is ≥25%. Results: Forty seven patients were accrued from July 2009 through June, 2010, 49 biopsies were performed (soft tissues n=31, liver n=8, bone n=3, lung n=1, adrenal gland n=1), 6 samples had poor cellularity and 3 failed array QC. There was no patient recall, hospitalization or emergency room visit due to biopsy procedure. The median time from biopsy to genomic prediction result was 5 days (range 3-7). Twenty three (57%) patients had positive result for at least 1 predictor (5 were positive for 2) and 20 are receiving therapy (3 withdraw or progressed before therapy began). Responses as of June 2010; Target index arm (n=9): 5 PD (progressive disease), 4 SD (3 stable disease at 8 weeks 1 SD at 16 weeks); SRC Pathway arm (n=5): 3 PD, 2 SD at 8 weeks; Cell line predictor arm (n=6): 2 PD, 1 SD at 8 weeks, 3 not yet reached response evaluation. None of the 3 predictive marker arms have met early stopping yet and accrual is ongoing. Conclusion: Gene-expression signature based patient selection for targeted therapy is feasible and FNA biopsies of metastatic lesions for genomic testing are safe. Updated efficacy results will be reported. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P6-14-06.

Background: P-and A-based regimens are widely used as PST and ACT of early BC. It is unknown whether the efficacy of such combinations is affected by their sequence of administration. The purpose of this retrospective analysis was to assess the clinical outcome of two different sequences of P and A in pts receiving PST and ACT. Methods: We analyzed 3,010 pts (1,414 PST and 1,596 ACT) with stage I — III BC treated between 1994 -2009. These pts were identified through our prospective online MDACC breast cancer database. Pts treated with trastuzumab, docetaxel or adriamycin/cyclophosphamide-regimen were excluded. Pathological complete response (pCR) was defined as no residual invasive disease in breast and ipsilateral axillary lymph nodes. Results: PST cohort included 1.071 pts (75%) with clinical stage I-IIIA and 343 (24.2%) stage IIIB-IIIC BC; 1,188 pts (84%) received the sequence P→A and 226 pts (16%) A→P. In the PST cohort 958 pts (67.7%) and 161 (11.3%) had hormone-receptor (HR) positive (+) and HER-2 + BC, respectively. ACT cohort consisted of 1,503 pts (94%) with pathological stage I-IIIA and 93 (5.82%) stage IIIB-IIIC. 1,196 pts (75%) received the sequence P→A, and 400 pts (25%) A→P. In the ACT cohort 1,122 (70.3%) and 93 (3.9%) had HR+ and HER2+ BC, respectively. Both cohorts of pts were balanced by HR-status, clinical stage, and menopausal status. The combined p-values for ACT and PST cohorts were calculated and they were significant for both RFS (p=0.022) and OS (p=0.002). In univariate analysis the sequence A→P was associated with inferior outcome. In Cox multivariate analysis, after stratification for period at diagnosis and adjustment for age, clinical stage, HR status, grade, and LVI the A→P sequence administered as PST was associated with significantly higher risk of relapse (HR 1.49; CI 1. 10-2.03; p=0.01) but not death (HR 1.28; CI 0.90-1.84; p=0.17). In the ACT cohort, the Cox multivariate analysis, after stratification for age, HR status, HER2 the sequence A→P was significantly associated with higher risk of death (HR 2.02; CI 1.33-3.06; p=0.001) but not relapse (HR 1.21; CI 0.82-1.79; p=0.33). Conclusions: The P→A sequence compared to the A→P sequence is associated with lower risk of relapse and death in PST and ACT, respectively. This retrospective analysis is hypothesis-generating and should lead to a prospective randomized trial to compare the two sequences as PST or ACT in locally advanced or early BC, respectively. Primary Systemic Therapy (N=1,414) Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P5-10-02.

Bien que le benefice de la chimiotherapie adjuvante dans le cancer du sein soit etabli, son effet absolu reste modeste. Il y a donc un besoin d’identifier des biomarqueurs qui permettraient de selectionner des patientes beneficiant de ce traitement. Malgre des recherches intenses dans ce domaine, aucun biomarqueur n’a ete demontre capable de predire le benefice de la chimiotherapie adjuvante dans la pratique quotidienne. Ceci peut etre explique en partie par le fait que l’entite «cancer du sein» comprendenre alite plusieurs maladies. Dans cet article, nous discuterons pourquoi l’approche de la recherche actuelle fondee sur un seul biomarqueur pour l’ensemble de la population est limitee dans le contexte d’une heterogeneite inter-tumorale. Nous proposerons ensuite trois solutions pour ameliorer les resultats actuels dans ce domaine, i. e. se concentrer sur un nombre limite de biomarqueurs dans une seule sous-classe homogene, evaluer des tests multi-marqueurs, et utiliser des voies fonctionnelles pour predire l’efficacite d’un medicament donne.

Introduction: To overcome limitations of previous multigene prognostic classifiers, we propose a new dynamic predictor. Our classifier does not use a single universal training cohort and an identical list of informative genes to predict the prognosis of new cases but a case-specific predictor is developed for each test case. Methods: We analyzed gene expression data from 3,534 breast cancers with clinical annotation including survival. For each test case we select a case-specific training subset including only molecularly similar cases and a case-specific predictor is generated. This method yields different training sets and different predictors for each new patient but preserves the independence of model building and validation. The model performance was assessed in leave-one-out validation and also in 325 independent cases. Results: Prognostic discrimination was high for all cases (n = 3,534, HR = 3.68, p = 1.67E-56). The dynamic predictor showed higher overall accuracy (0.68) than genomic surrogates for the 21-gene predictor (0.64), the 97-gene genomic grade index (0.55) or the 70-gene prognostic signature (0.41). The dynamic predictor was also effective in triple-negative cancers (n = 427, HR = 3.08, p = 0.0093) where the above classifiers all failed. Validation in independent patients yielded similar classification power (HR = 3.57). Discussion: We developed a new method to make personalized prognostic prediction using case-specific training cohorts. The dynamic predictors outperform static models developed from single historical training cohorts and they also predict well in triple-negative cancers. The dynamic classifier is available on line at http://www.recurrenceonline.com/?q = Re_training. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P3-05-05.

The reverse Monte Carlo structural modeling method and tight-binding molecular dynamics simulations were applied in order to investigate the existence of small bond angles (like those in triangles and squares) in amorphous silicon networks. The influence of small bond angles on the electronic density of states was also analyzed. It is shown that the presence of a small number of bond angles around 75° is necessary for a proper reproduction of the experimental structure factor of vapor deposited amorphous silicon.

The existence of small bond angles (like those of triangles and squares) in amorphous silicon networks were studied by the tight-binding molecular dynamics method, by analyzing the statistical data of Si–Si–Si fragments inside large molecules, and also by the Reverse Monte-Carlo simulation method. The influence of small bond angles on the electronic density of states was revealed.

Background: Little is known about how the immune microenvironment of breast cancer evolves during disease progression. Immunological differences between primary and metastatic lesions may explain discordant results of clinical trials that showed low tumor response rates with immune checkpoint therapy in metastatic breast cancer but high rates of pathologic complete response in early stage disease. The goal of this project was to examine TIL counts, PD-L1 protein expression and immune gene mRNA expression in primary tumors (P) and metastatic lesions (M). Methods: FFPE blocks of primary breast cancers and core needle biopsies of matching distant metastases from 54 patients (n=104 samples). TIL count was assessed on H&E slides for 39 paired tissues and is reported as % of TIL in the stroma. PD-L1 protein expression was detected with immunohistochemistry (IHC, E1L3N antibody) in 36 pairs, samples with > 1% cells showing staining were considered PD-L1 positive which was determined separately for tumor and stroma. The expression of 770 immune-related genes was measured using the Nanostring PanCancer Immune Gene Panel in 31 P and 17 M, including 10 paired cases. Genes were organized into 14 immune cell type (total T, Th1, Treg, Total CD8, exhausted CD8, Cytotoxic T, B, NK, NK-CD56, Mast cell, CD45, Dendritic cell, macrophage, neutrophil) and 22 immune function metagenes. Differences in mean expression in P and M were assessed using Fisher exact and Mann-Whitney tests without adjustment for multiple comparisons due to overlap in metagene membership. Results: Mean TIL counts (14% vs 20%, p=0.026) and stromal PD-L1 positivity by IHC (14% vs 54%, p=0.004) were significantly lower in M. PD-L1 positivity in tumor cells was similar (25% in M vs 42%, in P p=0.14). The total TIL gene expression score (2.48 vs 2.8 p=0.018) and all immune cell type metagenes, except neutrophils, had lower absolute expression levels in M. The relative abundance of neutrophils (0.035 vs -0.38, p=0.0001) and macrophages (0.62 vs 0.38 p=0.0013) increased in M. Among the 22 immune function metagenes, T, B and NK cell functions, cytotoxicity, chemokine and TNF superfamily expression, regulation and pathogen defense were significantly lower in M and none showed significantly increased expression in M. Pro-inflammatory/immune-activating cytokines of IL-6, CCL-5,-12,-19,-22 and CXCL-5,-9,-10,-11 were all significantly lower in M. The greatest drop was seen for CXCL-9 (2243 vs 422, p Conclusions: Breast cancer metastases exist in an attenuated immune microenvironment. Most immune cell subtypes, immune functions, and immune-associated gene expression are lower in M compared to P, consistent with immune escape. Metastatic lesions have higher relative abundance of macrophages and neutrophils, which suggest new therapeutic opportunities. Citation Format: Szekely B, Bossuyt V, Li X, Baine M, Silber A, Sanft T, Hofstatter E, Mougalian S, Baghwagar S, Neumeister V, Pelekanou V, Hatzis C, Pusztai L. Immunological differences between primary and metastatic 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 PD6-02.

Background Breast cancer is currently classified in 3 groups based on estrogen receptor alpha (ER) and human epidermal growth factor receptor 2 (HER2/ERBB2) gene expression: one basal-like (ER-ERBB2-), one HER2−enriched (ERBB2+) and one luminal (ER+). Yet, in transcriptome-based classifications, ER-ERBB2+ group partially overlaps with more recently defined ER-AR+ (androgen receptor positive) group. This type was named molecular apocrine, in reference to the histopathologically characterized apocrine carcinomas (H-Apo), in which a marked activation of AR signaling was demonstrated with a distinct proteomic signature. H-Apo tumors correspond to 1% of invasive breast carcinomas and are clearly morphological distinct from other AR+ tumors. However, no specific H-Apo transcriptome signature has been reported for this sub-group. In an effort to better characterize those tumors, we have performed a meta-analysis of genomic data, focusing on the ER- AR+ breast subset. Samples and Methods: Chips were from Affymetrix array generations HG-U133. 258 profiles were unpublished and 1145 were from published or in press data. Gene expression was carried out after GC-RMA normalization. Unsupervised hierarchical clustering and other statistical analysis were performed with R software. Results: 160 of the 1403 investigated tumors were ER-AR+. An unsupervised hierarchical clustering clearly identified a small subgroup of 14 closely tumors expressing high transcripts levels of PIP, HPGD, ACSM1, AR, SDR5A1, HS3DB1. This profile was very similar to the proteomic signature previously described for the H-Apo tumors. In addition, the pathology report, although available only for 4 of those14 tumors, described them as typical apocrine carcinomas. Taken together, these data suggested that this cluster was the H-Apo subgroup. Unexpectedly, when using the transcriptomic PAM50 classification, 13 were classified as Luminal and only 1 as HER2−enriched, although the 14 tumors were all ER-negative. CGH analysis with Agilent 244K chips was carried out with 25 ER- AR+ tumors, of which 5 were H-Apo carcinomas. Importantly, those 5 H-Apo tumors exhibited fewer DNA lesions than the other ER-AR+ apocrine tumors (17% copy number alterations in H-Apo group versus 41%, p=0.02). More CGH data are currently under investigations and will be discussed. Discussion: The histopathologically characterized apocrine carcinomas (H-Apo) display transcriptomic signs of active androgen metabolism and fewer DNA lesions than others molecular apocrine tumors. This could suggest that molecular apocrine and H-apocrine tumor derive from the same cell of origin, but that only H-Apo retains morphological apocrine features, possibly due to the presence of fewer genetic lesions. In any case, the prominent androgen signaling activation warrants functional assays of anti-androgen in these breast cancer subtypes. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P3-05-01.

MCGR (Monte Carlo determination of G(r)) is an inverse method for determining the pair correlation function from the structure factor. In this paper we describe the algorithm in detail, including recent developments such as the possible application of coordination constraints, Gaussian smoothing of peak shapes and the ability to subtract a (quadratic) polynomial background. We will illustrate how these have important practical implications for the way that experiments can be performed. Examples of the applications of MCGR to liquids/glasses and crystalline materials will be given.

This paper briefly describes the reverse Monte Carlo (RMC) method of structural modelling and its applications to studies of fast ion conducting glasses and (nonconducting) polymers. It then reviews the need for similar studies of polymer electrolytes, the prospects for such studies and the problems that are being encountered.

A simple numerical method is proposed whereby the relative information content of different total structure factors can be determined, whether obtained by neutron diffraction with isotopic substitu...