Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Cytoplasmic HAX1 Is an Independent Risk Factor for Breast Cancer Metastasis

Cytoplasmic HAX1 Is an Independent Risk Factor for Breast Cancer Metastasis Hindawi Journal of Oncology Volume 2019, Article ID 6375025, 13 pages https://doi.org/10.1155/2019/6375025 Research Article Cytoplasmic HAX1 Is an Independent Risk Factor for Breast Cancer Metastasis Alicja Trebinska-Stryjewska, Lukasz Szafron, Alina Rembiszewska , Maciej Wakula ,SylwiaTabor , Renata Sienkiewicz, Joanna Owczarek , Anna Balcerak , Anna Felisiak-Golabek ,and EwaA.Grzybowska Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena , - Warsaw, Poland Correspondence should be addressed to Ewa A. Grzybowska; ewag@coi.waw.pl Received 31 July 2018; Revised 20 February 2019; Accepted 7 March 2019; Published 10 April 2019 Guest Editor: Peramaiyan Rajendran Copyright © 2019 Alicja Trebinska-Stryjewska et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. HAX1 is an antiapoptotic factor involved in the regulation of cell migration and calcium homeostasis, overexpressed in several cancers, including breast cancer. It has been suggested that HAX1 is also implicated in metastasis. Herein we report the results of meta-analysis of HAX expression, based on publicly available data, which confirms its significant overexpression in breast cancer and demonstrates copy number gain and prognostic value of HAX overexpression for metastatic relapse in ER+ tumors. IHC analysis reported here also reveals its significant overexpre ssion in breast cancer samples from primary tumors, indicating significantly higher HAX1 protein levels in a group of patients who developed distant metastases in a disease course. Moreover, we demonstrate that HAX1 localization is important for the prediction of metastatic relapse and that cytoplasmic but not nuclear HAX1 is an independent risk factor for breast cancer metastasis. 1. Introduction dormant for as long as 20 years but eventually may start to proliferate. Late recurrences were observed in as much as 50% Breast cancer is the most common neoplasm and the primary of these cancers [4]. Basal cancers tend to metastasize early cause of cancer death in women [1]. Breast cancer mortality is (with a peak about 2-3 years aeft r diagnosis) and frequently almostexclusivelyduetometastaticdisease[2,3].Thecurrent [8, 9], but typically there is no recurrence aeft r 5 years in this 5-year survival for primary breast cancer is quite high (80- subtype. Distinct pattern of metastatic relapse in basal and 92%), but, despite the advances in diagnosis and treatment of luminal subtypes suggests different routes for metastasis. early breast cancer patients, about 20-40% experience distant Better molecular characteristic of the primary tumor organ metastases, for which the prognosis is signica fi ntly is crucial for a good prediction of the clinical outcome. worse [4–7]. Breast cancer is heterogeneous disease and Genetic tests such as MammaPrint (for luminal and basal a probability to develop metastases depends not only on cancers) [10] and Oncotype DX (luminal cancers only) [11] histopathological parameters (lymph node status, histologic were developed as a diagnostic tool to predict risk of breast grade, and tumor size) but also on molecular subtypes cancer metastasis, based on mRNA expression signature of defined roughly as basal-like, normal-like, HER2 −enriched, selected gene sets (70 and 21 genes, resp.). Quantification of and luminal A and luminal B, each of which has a different the risk of recurrence is especially important for selecting a prognosis and a pattern of recurrence. For luminal cancers subset of luminal patients who may benefit from additional (estrogen and/or progesterone receptor positive) the progno- chemotherapy andsparing thosewhowouldnot. sis is better due to a very eeff ctive adjuvant endocrine therapy Two other factors have both prognostic and predictive and the fact that the metastases appear late, oeft n many years values in breast cancer and are commonly used in risk- aer ft initial diagnosis. Disseminated tumor cells could stay assessment: urokinase plasminogen activator protein (uPA) 2 Journal of Oncology and its inhibitor (PAI-1). ELISA-based assay, developed to met the inclusion and exclusion criteria were selected: 23 assess the levels of both proteins in breast cancer tissues, consecutive patients with distant metastases and 23 consec- allows to stratify the patients with node-negative disease into utive patients without distant metastases within the follow- a low-risk group, with a good prognosis without adjuvant up period (comparative group). Assuming a power of 80%, chemotherapy and a high-risk group, with high expression of 𝛼=0.05, and normal distribution of data, a number of patients both markers, who may benetfi from chemotherapy [12, 13]. tested would allow to detect a difference of at least 0.43 Additionally, to monitor a response to the treatment and between group means (with a common standard deviation to assess the probability of metastasis, several blood-based equalto0.5)whenanalyzedwithMann-WhitneyUtest. biomarkers have been developed, including Human Epider- Patients’ age at the time of diagnosis, estrogen receptor mal Growth Factor Receptor 2 (HER2), Cancer Antigen 15- (ER) status, progesterone receptor (PGR) status, HER sta- 3 (CA 15-3, MUC1), and Carcinoembryonic Antigen (CEA) tus, tumor size (pTNM scale), node status (pTNM scale), [14]. clinical stage (AJCC Anatomic Stage Group), histological Herein we assessed the potential of HAX1 expression level grade (Nottingham Histologic Score system), histology, and in primary tumor samples as an independent prognostic fac- molecular subtype (based on routine immunohistochemical tor for breast cancer metastasis. HAX1 was rst fi characterized evaluation of ER, PGR, HER2 and Ki-67) were recorded. in 1997 as an antiapoptotic factor [15] and several reports Patients’ characteristics are shown in Table S1. confirm its involvement in the regulation of apoptosis [16– 18]. Additionally, HAX1 was implicated in the regulation of .. Follow-Up and Outcomes. Patients’ records were tracked cell motility [19–21] and calcium homeostasis [22]. from the time of surgery until May 2016. Information about HAX1 overexpression was reported in several cancers clinical outcomes (distant metastases confirmed by imaging [23–25], including breast cancer [26, 27], and its role in or histologic evidence, death from any cause) was retrieved metastasis was suggested in some reports [20, 28]. Sheng and from clinical records and The National Cancer Registry in Ni [28] reported that higher HAX1 expression was related to Poland. Distant relapse-free survival (DRFS) was defined a lower 10-year survival rate in breast cancer patients. according to STEEP system [34] as the time from surgical In this report we present data analysis which confirms intervention until the time of distant recurrence, death from significant HAX1 overexpression in breast cancer samples, any cause, or the last follow-up. Complete events in DRFS coinciding with high amplification of the HAX gene. More- analysis were distant metastasis or death, whichever came over, the analyses reveal significant difference between HAX1 rfi st. Overall survival (OS) was dene fi d as the time from levels in primary tumor samples between nonmetastatic surgery to the last follow-up (censored event) or to the time and metastatic groups of patients, indicating that HAX1 of death from any cause (complete event). may represent an independent risk factor for breast cancer metastasis. Additionally, it was demonstrated that IHC assay .. Immunohistochemistry. Immunohistochemical staining which takes into account protein localization may predict with a monoclonal mouse anti-HAX1 antibody (BD Bio- clinical outcome more precisely and with a higher strength sciences) or a control mouse IgG of the same subclass was than mRNA-based estimations. performed as described previously [27] on a set of represen- tative slides from formalin-fixed, paraffin-embedded breast 2. Materials and Methods tumors. HAX-1 expression was scored manually according to Ball et al. [35]. It was evaluated independently for nuclear and .. Study Group. Formalin-fixed, paraffin-embedded (FFPE) cytoplasmic staining. Light microscopy evaluation at 400x tissue samples were collected from breast cancer patients magnicfi ation was used to count 100 tumor cells within areas receivingsurgicalinterventionattheMariaSklodowska- of the strongest staining. Each nucleus and cytoplasm in a CurieInstitute,OncologyCenter, betweenJanuary2007 and given eld fi was assigned to an intensity category of 0 (absent), May 2007, aer ft informed consent. The study was approved 1 (weak), 2 (moderate), or 3 (strong). The percentage of cells by Ethics Committee from the Maria Sklodowska-Curie in each intensity category was determined as N0, N1, N2, Institute, Oncology Center, Warsaw, in accordance with the and N3, respectively. A distribution score (ID score) was then guidelines of the Helsinki Declaration of 1975, revised in calculated as 1983. Clinical data and histology reports for each patient were reviewed by two clinicians and a pathologist, respectively. (/N0∗0/+/N1∗1/+/N2∗2/+/N3∗3/) (1) ID= . De-identified patients data were accessed using MedStream Designer platform (Transition Technologies S.A.). Patients eTh ID score therefore ranged from 0 (absent staining in all were retrospectively analyzed and divided into two groups: cells) to a maximum 3 (100% cells having a staining intensity with and without distant metastases within the follow-up of 3). eTh values of total HAX1 staining were obtained by period of 9 years. Inclusion criteria for the study were as adding nuclear and cytoplasmic ID scores for each sample. follows: invasive breast cancer stages I-III, the absence of distant metastasis at the time of surgery, and the presence of tumor tissue in FFPE confirmed by a pathologist. Exclusion . . Immunofluorescence. MCF7 (ATCC), MDA-MB-231 criteria were as follows: previous history of breast cancer, (ATCC), and HeLa (ATCC) human cell lines were used in previous or simultaneous history of any other malignances, the experiments. All cell lines were authenticated by Eurofins and neoadjuvant chemotherapy. A total of 46 patients who Genomics (Germany). Cells were grown in Dulbecco’s Journal of Oncology 3 Modified Eagle Medium supplemented with 10% fetal bovine Survival analysis of patients with metastatic relapse serum (Thermo Fisher Scientific). Immunouo fl rescence information was also performed using Breast Cancer Gene- was performed as described previously [36] with primary Expression Miner v4.1 (bcGenExMiner v4.1) [33]. Patients anti-HAX1 antibody (rabbit, 1:100, eTh rmo Fisher Scienticfi ) were split into two groups according to gene’s expression and secondary goat anti-rabbit Alexa Fluor 594 antibody median and Kaplan-Meier survival curves were plotted for (1:500, er Th mo Fisher Scientific). Cells were observed using each group. Breast Cancer Gene-Expression Miner v4.1 was the Zeiss LSM 800 confocal microscope. Images represent also used in targeted prognostic analysis of HAX gene single Z-stacks. Colocalization was quantified using ImageJ expression for all patients with metastatic relapse informa- JACoP plugin [37], for 3-5 independent elds fi of vision and tion. eTh results summarize univariate Cox scores (hazard approximately 25-60 cells per efi ld. ratios, p-values) for each cohort fulfilling the chosen criteria and all of these cohorts pooled together. eTh results were .. Database Gene Expression Analysis. HAX gene expres- presented in forest plot. Additionally, multivariate Cox scores sion in primary breast cancer compared to normal breast (adjusted on NPI/AOL) were calculated for HAX. tissue was analyzed using the Oncomine Platform (eTh rmo Fisher, Ann Arbor, MI) [38, 39]. HAX expression (RNAseq) .. Statistical Analysis. Statistical analysis was performed in breast cancer in relation to phenotypic variables was using SAS Enterprise Guide 7.11 (Copyright ©2015 by SAS explored in TCGA-BRCA cohort (data generated by the Institute Inc., Cary, NC, USA) and Stata software (Stata- TCGA Research Network: http://cancergenome.nih.gov/) Corp., College Station, TX, StataCorp LP.). GraphPad Prism using Xena Functional Genomics Explorer (https://xena- version 6.07 for Windows (GraphPad Software, La Jolla, browser.net/, 2018) and in a set of microarray data using California USA, www.graphpad.com) was used to visual- bcGenExMiner (http://bcgenex.centregauducheau.fr, 2018) ize data. O’Brien-Castelloe approximation (SAS Enterprise [33]. Guide7.11) wasemployedfor statisticalpower andsamplesize analysis for Mann-Whitney U test. Baseline demographics, . . Database Copy Number Variation and Mutation Analysis. tumor characteristics, and types of treatment were compared Genomic alterations (mutations, gene amplification, and/or between the group of patients with distant metastases and the deletion) of HAX gene in breast cancer were assessed using comparative group using the Mann-Whitney U test for con- cBioPortal for Cancer Genomics (http://www.cbioportal.org/ tinuous and ordinal variables and by Pearson’s chi-squared index.do, 2018) [40, 41]. eTh following cohorts of invasive test for categorical variables. The Shapiro-Wilk W test was breast carcinoma were included: Breast Cancer (METABRIC) used to determine whether HAX1 protein levels measured by [29, 42], Breast Invasive Carcinoma (British Columbia) [43], immunohistochemistry were normally distributed. The asso- Breast Invasive Carcinoma (Broad) [44], Breast Invasive ciations between HAX1 immunoreactivity and progression, Carcinoma (Sanger) [45], Breast Invasive Carcinoma (TCGA, along with clinicopathological parameters, ER, PGR, and PanCancer Atlas) [30], Mutational profiles of metastatic HER2status,wereassessedusingtheMann-WhitneyUtestor breast cancer (France) [46], and the Metastatic Breast Cancer Kruskal-Wallis test, depending on whether the nominal vari- Project (Provisional, April 2018). eTh other cohorts were able had two or more categories. If significant, the Kruskal- excluded from the analysis due to patients overlapping Wallis test was followed by pairwise comparisons using or the difference in sample type (xenografts instead of the Mann-Whitney U test. Receiver operating characteristic primary tumor). Groups with shallow deletion (possibly (ROC) curve analysis was performed to determine the overall heterozygous deletion), diploid status, gain, or high-level test performance and to calculate possible cutoff points for amplicfi ation of HAX gene generated by GISTIC algorithm HAX1 protein levels. Optimal cutoff values were calculated [47] were compared for mRNA expression in METABRIC using the nearest to (0,1) method and the maximum value and TCGA cohorts. For the latter, HAX mRNA levels were of the Youden index. Kaplan-Meier survival analyses were also correlated with log2 copy number values using Pearson’s carried outfor overallsurvival(OS)and distantrelapse- correlation coefficient. free survival (DRFS). eTh log-rank test was used to evaluate HAX copy number variation in primary breast cancer theequalityofsurvivorfunctionforgroups with lowerand in comparison to normal tissues was also analyzed on the higher HAX1 expression categorized according to values Oncomine Platform using the following threshold values: p- obtained in the ROC curve analysis. The Cox proportional value 0.05, fold change ‘all’, and gene rank ‘top 5%.’ hazards model was used for univariate and multivariate analyses of patient survival depending on HAX1 expression, .. Database Survival and Prognostic Analysis. Survival anal- categorized as described above. In the multivariate survival yses of patients stratified according to HAX expression were analyses, HAX1 levels were assessed along with the following performed using KM Plotter (http://kmplot.com/analysis/, variables: PGR expression (categorization: positive vs. nega- 2018) [48]. HAX1 expression levels based on Affymetrix probe tive), clinical stage (I vs. II vs. III), histological grade (1 vs. ID 201145 at in35 cohortsofbreastcancerpatientsdeposited 2 vs. 3), and molecular subtype (luminal vs. others). Hazard in GEO database (Gene Expression Omnibus, NCBI) were ratios (HR) with 95% confidence intervals and p-values used. eTh general settings were as follows: patients split by were reported (Table 1). HAX gene expression retrieved median or by best cut-off; follow-up threshold: all; probe set from databases was compared in different subgroups using options: only JetSet best probe set; quality control: removing Student’s t-test and one-way Welch’s or Fisher’s ANOVA redundant samples and excluding biased arrays. followed by post-hoc Tukey-Kramer test. All tests used in this 4 Journal of Oncology Table 1: Multivariate Cox regression analysis of HAX1 levels in human breast cancers. Evaluation of HAX levels in the cell nuclei Variable name OS (/ ) DRFS (/ ) HR (95% CI), p HR (95% CI), p HAX1 (≤1.05 vs.>1.05) NS NS Histological grade (2 vs. 1) 6.06E+8 (2.26E+8-1.63E+9),<0.001 NS Clinical stage (III vs. I) 3.171 (1.017-9.884), 0.047 NS Evaluation of HAX levels in the cytoplasm Variable name OS (/ ) DRFS (/ ) HR (95% CI), p HR (95% CI), p HAX1 (≤1.02 vs.>1.02) NS . (.- . ), . Histological grade (2 vs. 1) 4.86E+8 (1.74E+8-1.36E+9),<0.001 NS Evaluation of HAX levels in both of the nuclei and the cytoplasm Variable name OS (/ ) DRFS (/ ) HR (95% CI), p HR (95% CI), p HAX1 (≤2.06 vs.>2.06) NS . (.  -. ), . Histological grade (2 vs. 1) 3.44E+8 (1.17E+8-1.01E+9),<0.001 NS eTh multivariate analysis of prognosis was carried out using the Cox proportional hazards model. Values before and after a slash (/) stand for the number of complete observations versus all observations, respectively. Only the results with p-values<0.05 are shown and those with p-values<0.05 for HAX1 expression are highlighted in italic type. HR and CI stand for the hazard ratio and confidence interval, respectively. OS: overall survival; DRFS: distant recurr ence-free survival; NS: a nonsignificant result (p ≥0.05). study were two-tailed and the significance level (alpha) was mutations in HAX sequence, one truncating E59X and two always set to 0.05. missense mutations, E39K and P259A. Majority of the iden- tified alterations comprised of high-level gene amplification which was detected in all 4 cohorts containing copy number 3. Results variationdata[29,30, 42,46] andrangedfrom5.16% to .. HAX Is Significantly Overexpressed in the Majority of 21.06% of all cases (average 16.01%) (Figure 2(a)). Addi- Analyzed Datasets of Breast Cancer Primary Tumor Samples. tionally, low-level HAX gene gain was identified in 36.81% HAX overexpression in primary breast cancer in comparison to 63.86% cases (average 45.92%) whereas shallow deletion to normal breast tissues was identified at mRNA [26, 27] and (possibly heterozygous deletion) was present in only 0% to protein level [27]. To further confirm this observation we per- 3.76% of patients (average 1.49%). Additionally, log-2 HAX formed analysis on breast cancer cohorts using the Oncomine gene copy number units were compared between blood, Platform [38, 39] and taking into account invasive breast breast, and invasive ductal and invasive lobular carcinoma in cancer samples (ductal and lobular). HAX gene expression in TCGA-BRCA cohort using the Oncomine Platform and were invasive primary tumor was significantly elevated compared foundtobeelevatedfor bothinvasive ductalcarcinoma(fold to normal tissue in 16 out of 19 analyses (Figure 1(a), legend in change: 1.273, p=9.26E-135, gene rank: top 1%, Figure 2(b)) Figure S1). Detailed analyses for ductal and lobular carcinoma and invasive lobular carcinoma (fold change: 1.297, p=1.13E- in selected datasets confirmed these conclusions (Figures 22, gene rank: top 2%, Figure 2(c)). 1(b)–1(e)). Two cohorts analyzed using cBioPortal, METABRIC [29], HAX expressioninbreastcancerinrelationtopheno- and TCGA-BRCA [30] contained gene expression informa- typicvariableswasassessedinasetofmicroarraydatausing tion so it was possible to relate HAX gene copy number Breast Cancer Gene-Expression Miner v4.1 (bcGenExMiner) with HAX mRNA level. In both cohorts mRNA expression [33]. This analysis revealed that HAX expression correlates differed significantly between putative groups with diploid positively with grade (Figure 1(f)), confirming our previous DNA content and HAX gene gain or amplification (Figures results, obtained on a small group of patients [27]. It was 2(d) and 2(e)). In TCGA cohort HAX1 log2 copy number also observed that HAX expression differs significantly values showed a moderate positive correlation with mRNA inmolecularsubtypesofbreastcancer, withthehighest expression (Pearson’s r=0.656, p<0.0001, Figure 2(f)). expression in basal and luminal B subtypes, associated with more aggressive neoplasm (Figure 1(g)). .. HAX Overexpression Is Associated with Cancer Relapse and Has Prognostic Impact on ER+ Subset. Survival analyses .. HAX Gene Copy Number Is Altered in Breast Can- of breast cancer patients stratified according to HAX expres- cer Patients. Analysis of HAX gene in 7 cohorts of inva- sion were performed using KM Plotter and microarray data sive breast carcinoma patients using cBioPortal for Cancer from 35 breast cancer cohorts from GEO (Gene Expression Genomics [40, 41] revealed HAX altered status in 15% Omnibus, NCBI). RFS (relapse-free survival) analysis includ- (549/3655) of sequenced cases. Only three patients had ing 3,951 patients showed a statistically significant difference Journal of Oncology 5 Comparison of HAX1 Across 19 Analyses Overexpression Median Rank p-Value Gene 1 5 10 25 25 10 5 1 1404.5 6.82E−4 HAX1 Not measured 12345672 8 9 10 11 12 13 14 15 16 17 18 190 (a) HAX1 expression in Zhao Breast HAX1 expression in TCGA Breast HAX1 expression in TCGA Breast (fold change 2.508, p=8.70E-4) (fold change 1.319, p=3.01E-17) (fold change 1.372, p=1.79E-9) 1.5 1.5 1.0 1.0 1.0 0.5 0.5 0.5 0.0 0.0 0.0 −0.5 −0.5 −0.5 −1.0 −1.0 −1.0 −1.5 −2.0 −1.5 −1.5 12 12 breast invasive ductal breast invasive ductal breast invasive lobular (n=3) (n=61) (n=61) breast carcinoma breast carcinoma breast carcinoma (n=38) (n=389) (n=36) (b) (c) HAX1 expression in Gluck Breast HAX1 expression in Curtis Breast HAX1 expression in Curtis Breast (fold change 1.349, p=0.005) (fold change 1.309, p=4.08E-35) (fold change 1.283, p=1.38E-17) 1.0 3.5 3.0 3.0 2.5 0.5 2.5 2.0 2.0 0.0 1.5 1.5 −0.5 1.0 1.0 0.5 −1.0 0.5 0.0 −1.5 −0.5 0.0 12 12 breast invasive breast invasive ductal breast invasive lobular (n=4) breast carcinoma (n=144) breast carcinoma (n=144) breast carcinoma (n=154) (n=1556) (n=148) (d) (e) Box plot of HAX1 expression Box plot of HAX1 expression according to according to SBR PAM50 subtypes TCGA (HAX1 expression) <0.0001 <0.0009 −2 −2 −4 −4 F=11.06 −6 −6 Ｊ<0.0001 Ｊ = 0.0047 Ｊ < 0.0001 SBR1 SBR2 SBR3 (No:) (546) (1431) (1316) (No:) (1060) (780) (1468) (1013) (681) PAM50 subtype (f) (g) Figure 1: HAX1 is overexpressed in primary breast tumor in comparison to normal breast tissue. (a-e) HAX expression in invasive breast cancer (ductal and lobular) in comparison to normal breast tissue assessed in publicly available datasets on the Oncomine Platform. (a) Comparison of HAX overexpression across 19 analyses. Dataset legend in Figure S1. (b-e) HAX overexpression in selected datasets [29– 32]. Differences between groups were assessed by Student’s t-test and results with p-values <0.05 were considered significant. (f) HAX expression in breast cancer samples stratified according to grade (Scarff-Bloom-Richardson grade, SBR) analyzed using bcGenExMiner. (g) HAX expression in breast cancer samples stratified according to molecular subtype (PAM50 classification) in a set of microarray data analyzed using bcGenExMiner [33] (left panel) or RNAseq TCGA-BRCA data (right panel). Differences between groups in (f) and (g) were assessed by Welch’s or Fisher’s ANOVA followed by post-hoc Tukey-Kramer test. Basal-like (n=98) HER2-enriched (n=58) Luminal A (n=231) Luminal B (n=127) Normal-like (n=8) log2 median-centered ratio HAX1 mRNA level log2 median-centered ratio HAX1 mRNA level log2 median-centered intensity log2 median-centered ratio Basal-like HER2-E Luminal A Luminal B Normal breast-like log2 median-centered ratio HAX1, gene expression RNAseq log2 median-centered intensity Illumina Hiseq [unit: log2(norm_count+1) 6 Journal of Oncology 20% 15% 10% HAX1 copy number in TCGA Breast HAX1 copy number in TCGA Breast invasive ductal breast carcinoma invasive lobular breast carcinoma 5% 2.0 1.0 1.5 0.5 1.0 0.5 0.0 0.0 −0.5 −0.5 1 23 123 blood breast invasive ductal blood breast invasive lobular Mutation breast carcinoma (n=702) (n=111) (n=702) (n=111) breast carcinoma Amplification (n=693) (n=71) (a) (b) (c) METABRIC (HAX1 expression) TCGA (HAX1 expression) <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 6 14 <0.0001 <0.0001 TCGA (CNV vs mRNA) <0.0001 12 −2 −2 R=0.656 (95%CI, 0.620-0.689) F=121.30 F=230,54 −2 Ｊ<0.0001 −4 Ｊ<0.0001 Ｊ<0.0001 −4 −6 −4 −1 0 1 2 3 4 HAX1, log2 copy-number values HAX1, Putative copy-number alterations HAX1, Putative copy-number alterations (e) (f) (d) Figure 2: HAX gene copy number is altered in breast cancer patients. (a) Alterations in HAX gene analyzed in 7 cohorts of invasive breast carcinoma patients using cBioPortal for Cancer Genomics. (b-c) HAX gene copy number in TCGA-BRCA data from the Oncomine Platform for (b) invasive ductal carcinoma and (c) invasive lobular carcinoma compared to blood and normal breast tissue. (d-e) Comparison of HAX expression in primary breast cancer samples in relation to DNA copy number in (d) METABRIC cohort [29] and (e) TCGA-BRCA cohort [30]. Differences between groups were assessed by Fisher’s ANOVA followed by post-hoc Tukey-Kramer test. (f) Correlation of HAX expression and log2 copy number values in TCGA-BRCA cohort (Pearson’s correlation coefficient). in survival, favoring patients with lower HAX expression significance was detected for ER+ (n=2,061) but not ER- regardless of whether patients were split by median (HR=1.37, (n=801) subgroup of patients (HR=1.18, 95% CI, 1.00-1.39, 95% CI, 1.22-1.52, log-rank p=2.2E-08, Figure 3(a)) or best log-rank p=0.044 and HR=1.12, 95% CI, 0.89-1.40, log-rank cutoff value (HR=1.42, 95% CI, 1.27-1.58, log-rank p=3.6E-10, p=0.33, resp., Figure 3(a)). FDR=1%). OS (overall survival) analysis in 1,402 patients also Prognostic analysis was carried out using bcGenExMiner indicated statistically significant more favorable prognosis for [33]. Targeted prognostic analysis for HAX in a group patients with lower HAX expression, but only if patients of ER-positive patients with metastatic relapse information were split by best cutoff value (HR=1.41, 95%CI, 1.12-1.77, log- (n = 2,822) revealed statistical significance (HR=1.15, p- rank p=0.0034) and at the expense of false discovery rate value=0.0008) for HAX expression level in the pooled (FDR=50%) (Figure S2A). cohort (Figure 3(b), left panel). Additionally, to evaluate In breast cancer ER status is one of the most important independent prognostic impact of HAX in ER+ patients prognostic and predictive factors. er Th efore, RFS analysis relative to the well-established breast cancer prognostic wasperformed usingKMPlotteronsubgroupsofbreast indexes, including Nottingham Prognostic Index (NPI) [49] cancer patients with different ER status, using median value and Adjuvant! Online (AOL) [50], adjusted Cox proportional of HAX expression to avoid high values of FDR. Statistical hazards model was used, revealing statistical significance for Shallow Deletion (n=25) Diploid (n=777) Gain (n=701) Amplification (n=401) Shallow Deletion (n=17) Diploid (n=273) Gain (n=686) Amplification (n=92) HAX1, mRNA Expression Z-Scores (Illumina Human v3 microarray) Alteration Frequency Breast (METABRIC) Breast (TCGAPanCan) e MBC Project BRCA (INSERM2016) Breast (BCCRC 2012) Breast (Broad 2012) Breast (Sanger) log2 copy number units HAX1, mRNA Expression Z-scores (Illumina HiSeq_RNASeqV2 syn4976369) log2 copy number units HAX1, mRNA Expression Z-Scores (Illumina HiSeq_RNASeqV2 syn4976369) Journal of Oncology 7 Total ER+ ER- HR = 1.12 (0.89 - 1.4) HR = 1.37 (1.22 − 1.52) HR = 1.18 (1 - 1.39) log-rank P = 0.33 log-rank P = 0.044 log-rank P = 2.2e−08 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 Time (months) Time (months) Time (months) Expression Expression Expression low low low high high high (a) ER+ ER- (b) Figure 3: HAX overexpression is higher in patients with breast cancer relapse and has prognostic impact on ER+ subset of patients. (a) RFS analysis for total number of patients and subsets with different ER status (left: ER+; right: ER-). Patients were split into groups with high and low HAX expression (based on microarray data, split by median). Kaplan-Meier estimates were generated in KM Plotter online tool for all data available for 2017 (merged datasets). Probability of cancer relapse is plotted against time. (b) Forest plots estimating prognostic impact of HAX expression in ER+ (n = 2,822) and ER- (n = 1,072) subsets of patients with metastatic relapse information (bcGenExMiner). Values in columns represent summarized univariate Cox scores (p-values, hazard ratios) for each cohort fulfilling the chosen criteria and for pooled cohorts. MR: metastatic relapse. HAX expressionadjustedonAOL(HR1.27, 95%CI1.06-1.52, were consistent with the previous RFS analyses, showing p-value: 0.0108, 12 cohorts, 382 patients, 101 metastases). signicfi ant difference for ER+ group of patients and the lack The same analysis performed for patients with negative of significance in ER- group of patients (Figure S2C). ER status (n=1,072) revealed the lack of statistical signif- icance (p-value=0.3853 for the pooled cohort) and even . . Cytoplasmic HAX Levels Are Significantly Higher in the the tendency for better prognosis associated with HAX1 Primary Tumor of Breast Cancer Patients Who Experience overexpression (Figure 3(b), right panel). Distant Metastasis during the Disease Course. 46 breast can- Prognostic analysis performed for HAX expression cer patients who were free of distant metastasis at the time regardless of ER status (n=3,924) indicated significance, but of surgery and received no neoadjuvant therapy were retro- bordering on the 0.05 threshold (HR=1.07, 95% CI, 1.00- spectively analyzed for HAX1 protein levels (cytoplasmic and 1.14, p-value: 0.0432, Figure S2B). Additionally, KM curves nuclear) in primary tumors by immunohistochemistry. Half for metastatic relapse-free survival (MRFS) were plotted in of the analyzed group developed distant metastasis during a bcGenExMiner for each group of patients with metastatic follow-up period of 9 years. Cytoplasmic HAX1 protein levels relapse information (all patients, ER+, ER-), and the results were significantly elevated (p=0.0003) in the group of patients Probability 0.0 0.2 0.4 0.6 0.8 1.0 Probability 0.0 0.2 0.4 0.6 0.8 1.0 Probability 0.0 0.2 0.4 0.6 0.8 1.0 8 Journal of Oncology p=0.0003 p=0.0761 p=0.0093 4 4 3 3 2 2 1 1 0 0 −1 −1 −1 distant distant distant distant distant distant metastasis-free metastasis metastasis-free metastasis metastasis-free metastasis (n=23) (n=23) (n=23) (n=23) (n=23) (n=23) (a) (b) (c) p=0.0166 distant metastasis distant metastasis-free p=0.0154 p=0.1335 p=0.0844 p=0.0379 p=0.0626 HAX1 IgG −1 −1 grade 1 grade 2 grade 3 grade 1 grade 2 grade 3 (n=5) (n=23) (n=16) (n=5) (n=23) (n=16) (d) (e) (f) Figure 4: HAX1 protein level in primary tumors stratified according to selected clinical and histological factors (presence of distant metastases, tumor grade). (a-d) HAX1 protein levels in the primary tumor were quantified from IHC data in distant metastasis-free versus distant metastasis group for (a) cytoplasmic, (b) total, and (c) nuclear HAX1 staining. (d) Representative images of HAX1 IHC and negative isotype control for patients from metastasis-free versus distant metastasis group.×40 objective, bar: 100𝜇m. (e) Cytoplasmic and (f) total HAX1 staining in breast cancers stratified according to tumor grade (grades 1-3). Results for individual patients and median and interquartile range for each group are shown. Differences in HAX1 protein levels between groups were assessed by the Mann-Whitney U test and results with p-values<0.05 were considered significant. with distant metastasis (median 1.50, mean±SD 1.48±0.92, .. High Cytoplasmic and Total HAX Protein Levels in Breast 95% CI of the mean 1.08-1.87) compared to the group with no Cancer Cells Are Risk Factors for Distant Metastasis and Death. distant metastasis (median 0.40, mean±SD 0.50±0.54, 95% To ascertain if HAX1 protein levels in primary tumor can CI of the mean 0.26-0.73) (Figure 4(a)). Total HAX1 staining be used as a prognostic factor in breast cancer, we analyzed wasalsohigherinthe distantmetastasispatient group, follow-up patient data and recorded time to distant recur- although the eeff ct was less prominent (metastasis: median rence and/or time to death from any cause for all 46 patients. 1.94, mean±SD 1.93±0.85, 95% CI of the mean 1.56-2.29 vs. The total follow-up time was 9 years; 61% of the patients metastasis-free: median 1.37, mean±SD 1.22±0.76, 95% CI of had been followed for a minimum of 5 years. 23 patients the mean 0.89-1.55, p=0.0093) (Figure 4(b)). eTh opposite developed distant metastasis. 23 out of 46 patients were still effect, albeit not statistically significant, was observed for aliveattheendofthefollow-upperiod(18 in agroup with no nuclear HAX1 levels (metastasis: median 0.00, mean±SD distant metastasis and 5 in a group with distant metastasis). 0.45±0.60, 95% CI of the mean 0.19-0.71 vs. metastasis-free: Receiver operating characteristic (ROC) analysis was median 1.00, mean±SD 0.73±0.66, 95% CI of the mean 0.44- performed to define the best cutoff value of the HAX1 signal 1.01, p=0.0761) (Figure 4(c)). Representative images of the andtomeasuretheoveralltest performancewhich would typical staining in metastasis-free and metastatic groups are use HAX1 protein levels to predict breast cancer metastasis. presented in Figure 4(d). eTh analysis was done separately for cytoplasmic, total, eTh two analyzed groups were well matched, as patients’ and nuclear HAX1 immunohistochemical staining (Figures clinicopathological parameters and treatment did not differ 5(a)–5(c)). The highest value of area under the curve (AUC) significantly except for PGR status (Table S1). Analyses of was obtained for cytoplasmic HAX1: 0.7977 (95% CI 0.6628- HAX1 protein levels in groups stratiefi d according to known 0.9327, p=0.0005) (Figure 5(a)). eTh best cutoff points for prognostic factors showed that the values of the cytoplasmic cytoplasmic, nuclear, and total HAX1 were, respectively, 1.02 and total HAX1 signal were positively associated with tumor (sensitivity 0.65 and specificity 0.87), 1.05 (sensitivity 0.82 grade (Figures 4(e) and 4(f), resp.), but not other prognostic and specificity 0.48), and 1.49 (sensitivity 0.78 and specificity factors. 0.57). HAX1 cytoplasmic staining HAX1 total staining HAX1 cytoplasmic staining HAX1 nuclear staining HAX1 total staining Journal of Oncology 9 ROC of HAX1 cytoplasmic staining ROC of HAX1 total staining ROC of HAX1 nuclear staining 1.0 1.0 1.0 0.5 0.5 0.5 AUC (95% CI)=0,7977 (0,6628-0,9327) AUC (95% CI)=0,7212 (0,5742-0,8678) AUC (95% CI)=0,6456 (0,4835-0,8076) p=0,0907 p=0,0005 p=0,0102 0.0 0.0 0.0 0.0 0.5 1.0 0.0 0.5 1.0 0.0 0.5 1.0 1 - Specificity 1 - Specificity 1 - Specificity Sensitivity Sensitivity Sensitivity identity identity identity (a) (b) (c) HAX1 cytoplasmic staining HAX1 nuclear staining 110 HAX1 total staining p=0.0012 p=0.0074 p=0.3400 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 DRFS (months) DRFS (months) DRFS (months) cytoplasmic HAX1 ID score ≤1.02 total HAX1 staining ≤1.49 HAX1 nuclear staining ≤1.05 cytoplasmic HAX1 ID score >1.02 total HAX1 staining >1.49 HAX1 nuclear staining >1.05 HAX1 nuclear staining HAX1 total staining 110 110 HAX1 cytoplasmic staining 100 100 90 90 80 80 70 70 60 60 30 30 20 20 20 p=0.4456 p=0.0097 p=0.0134 10 10 0 0 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 OS (months) OS (months) OS (months) HAX1 cytoplasmic staining ≤1.02 HAX1 nuclear staining ≤1.05 HAX1 total staining ≤1.49 HAX1 cytoplasmic staining >1.02 HAX1 nuclear staining >1.05 HAX1 total staining >1.49 (f) (d) (e) Figure 5: HAX1 protein level in primary tumor is a risk factor for breast cancer progression. (a-c) Receiver operating characteristic analysis for (a) cytoplasmic, (b) total, and (c) nuclear HAX1 protein levels was performed to define the best cutoff values for subsequent survival analysis. Area under curve (AUC) with 95% CI and p-values for each ROC curve are shown. (d-f) Kaplan-Meier survival estimates for distant recurrence-free survival (DRFS) and overall survival (OS) in breast cancer patients according to proposed cutoff values of (d) cytoplasmic HAX1 protein levels≤1.02 (n=28) versus>1.02 (n=18), (e) total HAX1 protein levels≤1.49 (n=18) versus>1.49 (n=28), and (f) nuclear HAX1 protein levels≤1.05 (n=31) versus>1.05 (n=15). eTh log-rank test was used to evaluate the equality of survivor function for groups with lower and higher HAX1 expression and p-values<0.05 were considered significant. Cutoff points estimated from ROC curves were used in compared to 89% of patients in the group with cytoplasmic subsequent survival analyses by the Kaplan-Meier method. HAX1 protein levels>1.02. Overall survival analysis showed The log-rank test showed a significant difference favoring, that,attheendoffollow-up,64% ofpatientswerestillalivein for both distant recurrence-free survival (DRFS) and overall the group with a cytoplasmic HAX1 of≤1.02 compared to 28% survival (OS), patients with a cytoplasmic HAX1 ID score of patients in the group with cytoplasmic HAX1>1.02. Similar of≤1.02 (p=0.0012 and p=0.0134, resp., Figure 5(d)). 43% of results were observed for total HAX1 protein levels. Patients patients in the group with cytoplasmic HAX1 protein levels with total HAX1 protein levels ≤1.49 showed significantly ≤1.02 experienced distant metastasis/death within 9 years increased DRFS and OS compared to the group with a total Percent survival Percent survival Sensitivity Percent survival Percent survival Sensitivity Sensitivity Percent survival Percent survival 10 Journal of Oncology HAX1 Nuclei (DAPI) Merged Figure 6: HAX1 localization in breast cancer cell lines of different characteristics. Endogenous staining of HAX1 (red) and nuclei (DAPI, blue) in MCF7, luminal-like epithelial cells and MDA-MB-231, basal-like cells aer ft epithelial-mesenchymal transition. Bar: 20 𝜇m. HAX1 protein level of>1.49 (p=0.0074 and p=0.0097, resp., analysis to assess its eeff ct on metastasis. Database analysis Figure 5(e)). Nuclear HAX1 staining showed no prognostic on large group of patients conrfi med HAX overexpression in value for neither DRFS nor OS (Figure 5(f)). breast cancer samples, which tallies with the previous study Overall survival (OS) and distant recurrence-free survival by Luoetal. [26].Theanalysisrevealedalsothecorrelationof (DRFS) for 46 breast cancer patients were also evaluated HAX1 overexpression with tumor grade, which is consistent by 3 different univariate and multivariate analyzes, in which with our previous [27] and current IHC results. Additionally, the HAX1 protein expression, either cytoplasmic, nuclear, or HAX overexpression was shown to correlate with gene cumulative, was assessed by IHC (Table 1). We found out amplicfi ation. Although there were several studies reporting that elevated HAX1 levels in the cytoplasm emerged as an HAX overexpression in different types of malignancies, we independent, negative prognostic factor, associated with an showed for the first time that high HAX mRNA levels in increased risk of distant metastasis (HR 2.832, 95% CI 1.207- cancer cells could be a consequence of gene amplicfi ation, at 6.644, p=0.017). Correspondingly, the results obtained for least in breast cancer. Detailed analysis of HAX expression cumulative expression of HAX1 also showed its adverse effect in molecular subtypes demonstrated that the highest overex- on DRFS (HR 4.249, 95% CI 1.404-12.86, p=0.010). HAX1 pression was observed in basal and luminal B subtypes, which nuclear expression had no impact on survival. are more aggressive. Database analysis of HAX expression in correlation to . . HAX Localization Varies among Breast Cancer Cell Lines. metastasis revealed its signicfi ant prognostic value for lumi- Endogenous HAX1 protein was detected by immunouo- fl nal (ER+) subset while for ER-, despite high overexpression rescence in luminal-like MCF7 and basal-like MDA-MB- in basal cancers, the expression level had no prognostic value. 231 cell lines, revealing significant differences. In MCF7 cells This apparent paradox can be resolved on the basis of cellular HAX1 staining was mostly cytoplasmic, while in MDA-MB- localization. 231 HAX1 was also detected in the nuclei (Figure 6). Nuclear Our previous IHC analysis [27] indicated two dieff rent colocalization was calculated using ImageJ JACoP, showing localizationsofHAX1proteininbreastcancertumorsamples: a significant shift of Pearson’s correlation coefficient (PCC) cytoplasmic and nuclear. Nuclear localization of HAX1 was and two Mander’s overlap coefficients (M1, M2) from 0.101 also reported in cell lines [36] and rat tests [51]. Different (PCC), 0.207 (M1), and 0.116 (M2) in MCF7 cells to 0.467 localization may translate into different functionality and (PCC), 0.377 (M1), and 0.592 (M2) in MDA-MB-231 cells, different impact on tumor progression, as in case of Aurora respectively (p-values for PCC, M1, and M2: 0.0105, 0.0328, A kinase, where nuclear protein acquires kinase-independent and 0.0181, resp.). transactivating function, which enhances breast cancer stem cell phenotype [52]. u Th s, in this report, we have analyzed 4. Discussion HAX1 protein levels in the primary tumor of breast cancer Advancing on our previous study [27], in which we demon- patients divided into metastatic and nonmetastatic groups. strated HAX overexpression in breast cancer and its differen- IHC analysis enabled us to differentiate between cytoplas- tial localization (cytoplasmic and nuclear), we expanded our mic and nuclear localization of HAX1. Overall, our results MDA-MB-231 MCF7 Journal of Oncology 11 demonstrated that HAX1 protein level is significantly higher ausefultoolforestimating theprobability ofluminalbreast in metastatic group of patients, but this effect can be observed cancer dissemination. only for evaluations concerning cytoplasmic and total HAX1, while for nuclear localization it does not exist and the trend Data Availability is even opposite (less HAX1 in metastatic group). Clearly, the results for total HAX1 levels are inu fl enced by the cytoplasmic eTh data used to supportthe nfi dingsofthisstudy are subset, for which the difference is huge. available from the corresponding author upon request. uTh s, ourevaluationofHAX1 proteinlevelsandlocal- ization in the samples from metastatic versus nonmetastatic Conflicts of Interest groups of patients indicates a positive relationship between HAX1 cytoplasmic expression and the occurrence of a The authors declare no conflicts of interest. secondary tumor at distant locations in the course of the disease (opposite to the relations observed for Aurora A). Funding High cytoplasmic HAX1 level is associated negatively with progression-free survival and overall survival. Similar results This work was supported by the Polish National Science Cen- were obtained for total HAX1; however, ROC curve analysis ter Grants Nos. 2011/01/B/NZ1/03674, 2014/14/M/NZ1/00437, indicated a higher ability to identify patients at risk of and 2016/21/B/NZ2/03473. MedStream Designer (Transition progression when using cytoplasmic, not total HAX1 levels. Technologies S.A.) purchase was n fi anced by ONKO.SYS Accordingly, the experimental immunouo fl rescence project (Grant No. POIG.02.03.00-14-084/13) from the Polish results showing that HAX1 localization is more cytoplasmic National Centre for Research and Development. in luminal-like than basal-like cell lines can explain the apparent difference in HAX1 prognostic value for ER+ and ER- subsets. It seems plausible that, in basal cells, Acknowledgments despite the high HAX1 expression, nuclear localization Wewouldliketothank MariaZwierko,PhD forproviding of HAX1 prevents its prometastatic action, by assuming patients’ data from eTh National Cancer Registry in Poland. different functionality or simply by sequestering cytoplasmic HAX1. Alternatively, it seems plausible that nuclear HAX1 can block and/or sequester in inactive complexes some Supplementary Materials nuclear factor(s), specific to luminal cancers, whose action is linked to metastasis, probably by the regulation of Table S1: clinical and pathological characteristics of breast cancer patients. Figure S1: dataset legend to Oncomine transcription. u Th s, nuclear HAX1 would have protective meta-analysis of HAX expression in breast cancer (ductal effect (restricted to luminal-like cells), which would not and lobular). Figure S2: HAX overexpression is associ- be present in cells with cytoplasmic HAX1. Of note, HAX1 ated with metastatic relapse and survival (additional data). was shown to bind directly estrogen receptor [53], which (Supplementary Materials) is a prometastatic factor. Estrogen responsiveness is one of the main factors differentiating between the two cell lines used in experiments. However, since luminal patients are References treated with anti-estrogen therapy, some other factors may be [1] F. Bray, J. Ferlay, I. Soerjomataram et al., “Global cancer statistics involved. Further research should explain specific molecular 2018: GLOBOCAN estimates of incidence and mortality world- mechanisms of function for cytoplasmic and nuclear HAX1, wide for 36 cancers in 185 countries,” CA: A Cancer Journal for which contribute to metastasis. Clinicians,vol.68,no.6,ArticleID30207593,pp.394–424,2018. [2] C. L. Chaer ff and R. A. Weinberg, “A perspective on cancer cell 5. Conclusions metastasis,” Science,vol.331,no. 6024,pp.1559–1564,2011. [3] F.Cardoso,A.Costa,L.Nortonetal.,“1stInternational Overall, presented results indicate HAX overexpression, consensus guidelines for advanced breast cancer (ABC 1),” copy number gain, and prognostic impact on metastatic Breast,vol.21,no.3,ArticleID 22425534,pp.242-52,2012. breast cancer. Moreover, cytoplasmic but not nuclear HAX1 [4] X.H.Zhang,M.Giuliano,M.V.Trivedi,R. Schi,a ff ndC. K. demonstrated to be an independent, negative prognostic Osborne, “Metastasis dormancy in estrogen receptor-positive factor for breast cancer metastasis. breast cancer,” Clinical Cancer Research,vol.19, no.23, Article Multigene tests like MammaPrint and Oncotype DX have ID 24298069, pp. 6389–6397, 2013. a proved prognostic value, but they rely on expression on [5] Z. H. Li, P. H. Hu, J. H. Tu et al., “Luminal B breast cancer: mRNA level, which does not always accurately correspond to patterns of recurrence and clinical outcome,” Oncotarget,vol. protein level and does not consider the factor of protein local- 7, no. 40, Article ID 27542253, pp. 65024–65033, 2016. ization. It is possible that for some markers establishing the [6] R. L. Siegel, K. D. Miller, and A. Jemal, “Cancer statistics, 2015,” level of localized protein provides more accurate prediction CA: A Cancer Journal for Clinicians,vol.65,no.1,ArticleID than mRNA profiling. Further analysis is required to conrm fi 25559415, pp. 5–29, 2015. its prognostic value in the clinic, but assessing HAX1 protein [7] J. Ferlay, I. Soerjomataram, R. Dikshit et al., “Cancer incidence levels and localization in primary tumor samples may become and mortality worldwide: sources, methods and major patterns 12 Journal of Oncology in GLOBOCAN 2012,” International Journal of Cancer,vol.136, [22] E.Vafiadaki,D.A.Arvanitis,S.N.Pagakisetal.,“ea Th nti- no. 5, Article ID 25220842, pp. E359–86, 2014. apoptotic protein HAX-1 interacts with SERCA2 and regulates Its protein levels to promote cell survival,” Molecular Biology of [8] F. Bertucci, P. Finetti, and D. Birnbaum, “Basal breast cancer: the Cell (MBoC), vol. 20, no. 1, Article ID 18971376, pp. 306–318, A complex and deadly molecular subtype,” Current Molecular Medicine, vol. 12, no. 1, Article ID 22082486, pp. 96–110, 2012. [9] H. Kennecke, R. Yerushalmi, R. Woods et al., “Metastatic behav- [23] X.Li,J.Jiang,R.Yangetal.,“Expression of HAX-1incolorectal ior of breast cancer subtypes,” Journal of Clinical Oncology,vol. cancer and its role in cancer cell growth,” Molecular Medicine 28, no. 20, Article ID 20498394, pp. 3271–3277, 2010. Reports,vol.12,no.3,ArticleID26062578,pp.4071–4078, 2015. [10] M. J. van de Vijver, Y. D. He, L. J. van’t Veer et al., “A gene- [24] X. Li, T. Li, B. You et al., “Expression and function of HAX-1 in expression signature as a predictor of survival in breast cancer,” uman cutaneous squamous cell carcinoma,” Journal of Cancer, e New England Journal of Medicine,vol.347,no.25,Article ID vol. 6, no. 4, Article ID 25767605, pp. 351–359, 2015. 12490681, pp. 1999–2009, 2002. [25] H. Wu, J. Chen, Q. Wang et al., “Abnormal expression of HAX1 [11] S. Paik, S. Shak, G. Tang et al., “A multigene assay to predict is associated with cellular proliferation and migration in human recurrence of tamoxifen-treated, node-negative breast cancer,” hypopharyngeal squamous cell carcinoma,” Molecular Medicine e New England Journal of Medicine, vol. 351, no. 27, Article ID Reports, vol. 16, no. 4, Article ID 28791389, pp. 4664–4670, 2017. 15591335, pp. 2817–2826, 2004. [26] J. Hu, Z. Li, X. Li et al., “hSav1 interacts with HAX1 and [12] V. R. Jacobs, R. E. Kates, E. Kantelhardt et al., “Health eco- attenuates its anti-apoptotic effects in MCF-7 breast cancer nomic impact of risk group selection according to ASCO- cells,” International Journal of Molecular Medicine,vol.28, no. recommended biomarkers uPA/PAI-1 in node-negative pri- 3, Article ID 21567072, pp. 349–355, 2011. mary breast cancer,” Breast Cancer Research and Treatment,vol. [27] A. Trebinska, A. Rembiszewska, K. Ciosek et al., “HAX-1 138, no. 3, pp. 839–850, 2013. overexpression, splicing and cellular localization in tumors,” [13] F.Janicke,A.Prechtl,C.Thomssenetal.,“Randomizedadjuvant BMC Cancer, vol. 10, no. 76, Article ID 20196840, 2010. chemotherapy trial in high-risk, lymph node-negative breast [28] C. Sheng and Q. Ni, “Expression of HAX1 and Ki-67 in breast cancer patients identified by urokinase-type plasminogen acti- cancer and its correlations with patient’s clinicopathological vator and plasminogen activator inhibitor type 1,” JNCI: Journal characteristics and prognosis,” International Journal of Clinical of the National Cancer Institute,vol.93,no.12,ArticleID and Experimental Medicine,vol.8,no. 11,ArticleID26885017, 11416112, pp. 913-20, 2001. pp. 4–10, 2090. [14] R. Molina,J.M.Auge, J. M. Escudero et al., “Evaluation of tumor markers (HER-2/neu oncoprotein, CEA, and CA 15.3) [29] C. Curtis, S. P. Shah, S. F. Chin et al., “eTh genomic and tran- in patients with locoregional breast cancer: prognostic value,” scriptomic architecture of 2, 000 breast tumours reveals novel Tumor Biology,vol.31,no.3,ArticleID 20361287,pp.171–180, subgroups,” Nature,vol.486,no. 7403,Article ID 22522925,pp. 2010. 346–352, 2012. [15] Y. Suzuki, C. Demoliere, D. Kitamura et al., “HAX-1, a novel [30] N. Cancer Genome Atlas, “Comprehensive molecular portraits intracellular protein, localized on mitochondria, directly asso- of human breast tumours,” Nature,vol.490,no. 7418,Article ID ciates with HS1, a substrate of Src family tyrosine kinases,” e 23000897, pp. 61–70, 2012. Journal of Immunology, vol. 158, no. 6, Article ID 9058808, pp. [31] S. Gluck, J. S. Ross, M. Royce et al., “TP53 genomics predict 2736-44, 1997. higher clinical and pathologic tumor response in operable [16] Y. Han, Y. Chen, Z. Liu et al., “Overexpression of HAX-1 protects early-stage breast cancer treated with docetaxel-capecitabine cardiac myocytes from apoptosis through caspase-9 inhibition,” +/- trastuzumab,” Breast Cancer Research and Treatment,vol. Circulation Research,vol.99,no.4,ArticleID16857965,pp.415– 132, Article ID 21373875, pp. 781–791, 2012. 423, 2006. [32] H. Zhao, A. Langerod, Y. Ji et al., “Different gene expression [17] L. Cilenti, M. M. Soundarapandian, G. A. Kyriazis et al., patterns in invasive lobular and ductal carcinomas of the breast,” “Regulation of HAX-1 anti-apoptotic protein by Omi/HtrA2 Molecular Biology of the Cell (MBoC),vol.15,no.6,Article ID protease during cell death,” eJournalofBiologicalChemistry, 15034139, pp. 2523–2536, 2004. vol.279,no.48, pp.50295–50301,2004. [33] P. Jez ´ eq ´ uel, M. Campone, W. Gouraud et al., “bc-GenExMiner: [18] T. V. Sharp, H. W. Wang, A. Koumi et al., “K15 protein of Kaposi’s an easy-to-use online platform for gene prognostic analyses in sarcoma-associated herpesvirus is latently expressed and binds breast cancer,” Breast Cancer Research and Treatment,vol.131, to HAX-1, a protein with antiapoptotic function,” Journal of no.3,Article ID21452023,pp.765–775, 2012. Virology,vol.76,no.2,ArticleID 11752170,pp. 802–816,2002. [34] C. A. Hudis, W. E. Barlow, J. P. Costantino et al., “Proposal [19] P. J. Cavnar, E. Berthier, D. J. Beebe, and A. Huttenlocher, “Hax1 for standardized definitions for efficacy end points in adjuvant regulates neutrophil adhesion and motility through RhoA,” e breast cancer trials: the STEEP system,” Journal of Clinical Journal of Cell Biology, vol. 193, no. 3, Article ID 21518791, pp. Oncology,vol.25,no.15, ArticleID17513820, pp.2127–2132, 465–473, 2011. [20] A. G. Ramsay, M. D. Keppler, M. Jazayeri et al., “HS1-associated [35] E. Ball, J. Bond, B. Franc et al., “An immunohistochemical study protein X-1 regulates carcinoma cell migration and invasion of p16(INK4a) expression in multistep thyroid tumourigenesis,” via clathrin-mediated endocytosis of integrin 𝛼v𝛽6,” Cancer European Journal of Cancer,vol.43, no.1,Article ID 17046239, Research, vol. 67, no. 11, Article ID 17545607, pp. 5275–5284, pp.194–201,2007. [21] V. Radhika, D. Onesime, J. H. Ha, and N. Dhanasekaran, “G𝛼13 ´ [36] E. A. Grzybowska, V. Zayat, R. Konopinski et al., “HAX-1 is stimulates cell migration through cortactin-interacting protein a nucleocytoplasmic shuttling protein with a possible role in Hax-1,” eJournalofBiologicalChemistry,vol.279,no.47, mRNA processing,” FEBS Journal,vol.280,no. 1, ArticleID Article ID 15339924, pp. 49406–49413, 2004. 23164465, pp. 256–272, 2013. Journal of Oncology 13 [37] S. Bolte and F. P. Cordelier ` es, “A guided tour into subcel- [53] M.P.Walker,M.Zhang,T.P.Le, P. Wu,M.Laine, ´ and G. lular colocalization analysis in light microscopy,” Journal of L. Greene, “RAC3 is a pro-migratory co-activator of ER𝛼,” Microscopy,vol.224,no. 3,pp.213–232,2006. Oncogene, vol. 30, no. 17, pp. 1984–1994, 2011. [38] D.R. Rhodes,J.Yu, K.Shankeretal.,“ONCOMINE, acancer microarray database and integrated data-mining platform,” Neoplasia,vol.6,no.1, ArticleID15068665,pp.1–6,2004. [39] D. R. Rhodes, S. Kalyana-Sundaram, V. Mahavisno et al., “Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles,” Neoplasia,vol.9,no. 2, Article ID 17356713, pp. 166–180, 2007. [40] J. Gao, B. A. Aksoy, U. Dogrusoz et al., “Integrative analysis of complex cancer genomics and clinical profiles using the cBio- Portal,” Science Signaling,vol.6,no.269,ArticleID23550210, [41] E.Cerami,J.Gao,U.Dogrusozetal.,“ec Th Biocancergenomics portal: an open platform for exploring multidimensional cancer genomics data,” Cancer Discovery,vol.2,no.5,ArticleID 22588877, pp. 401–404, 2012. [42] B. Pereira, S. F. Chin, O. M. Rueda et al., “The somatic mutation profiles of 2, 433 breast cancers refines their genomic and transcriptomic landscapes,” Nature Communications,ArticleID 27161491, 2016. [43] S. P. Shah, A. Roth, R. Goya et al., “eTh clonal and mutational evolution spectrum of primary triple-negative breast cancers,” Nature,vol.486,no. 7403,ArticleID 22495314,pp. 395–399, [44] S. Banerji, K. Cibulskis, C. Rangel-Escareno et al., “Sequence analysis of mutations and translocations across breast cancer subtypes,” Nature, vol. 486, no. 7403, Article ID 22722202, pp. 405–409, 2012. [45] P. J. Stephens, P. S. Tarpey, H. Davies et al., “eTh landscape of cancer genes and mutational processes in breast cancer,” Nature, vol. 486, no. 7403, Article ID 22722201, pp. 400–404, 2012. [46] C. Lefebvre, T. Bachelot, T. Filleron et al., “Mutational profile of metastatic breast cancers: a retrospective analysis,” PLoS Med, vol. 13, no. 2, Article ID e1002201, 2016. [47] R. Beroukhim, G. Getz, L. Nghiemphu et al., “Assessing the significance of chromosomal aberrations in cancer: methodol- ogy and application to glioma,” in Proceedings of the National Academy of Sciences of USA, vol. 104, pp. 20007-12, 2007 Dec 11. [48] B. Gyorffy, A. Lanczky, A. C. Eklund et al., “An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients,” Breast Cancer Research and Treatment,vol.123,no.3, ArticleID 20020197, pp. 725–731, 2010. [49] M.H.Galea, R.W.Blamey,C.E.Elston,andI.O.Ellis,“eTh Nottingham prognostic index in primary breast cancer,” Breast Cancer Research and Treatment,vol.22,no.3,pp.207–219, 1992. [50] P. M. Ravdin, L. A. Simino,ff G. J. Davis et al., “Computer pro- gram to assist in making decisions about adjuvant therapy for women with early breast cancer,” Journal of Clinical Oncology, vol. 19, no. 4, Article ID 11181660, pp. 980–991, 2001. [51] E. Sarnowska, E. A. Grzybowska, K. Sobczak et al., “Hairpin structure within the 3’UTR of DNA polymerase beta mRNA acts as a post-transcriptional regulatory element and interacts with Hax-1,” Nucleic Acids Research,vol.35, no.16, ArticleID 17704138, pp. 5499–5510, 2007. [52] F. Zheng, C. Yue, G. Li et al., “Nuclear AURKA acquires kinase- independent transactivating function to enhance breast cancer stem cell phenotype,” Nature Communications,vol.7,no.10180, Article ID 26782714, 2016. MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Immunology Research Endocrinology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com BioMed PPAR Research Research International Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2013 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Neurology Research and Treatment Cellular Longevity Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Oncology Hindawi Publishing Corporation

Cytoplasmic HAX1 Is an Independent Risk Factor for Breast Cancer Metastasis

Download PDF
14 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/cytoplasmic-hax1-is-an-independent-risk-factor-for-breast-cancer-dW0SvNGjTU

References (55)

Publisher
Hindawi Publishing Corporation
Copyright
Statistical analysis was performed using SAS Enterprise Guide 7.11 (Copyright ©2015 by SAS Institute Inc., Cary, NC, USA) and Stata software (StataCorp., College Station, TX, StataCorp LP.). GraphPad Prism version 6.07 for Windows (GraphPad Software, La Jolla, California USA, www.graphpad.com) was used to visualize data. O’Brien-Castelloe approximation (SAS Enterprise Guide 7.11) was employed for statistical power and sample size analysis for Mann-Whitney U test. Baseline demographics, tumor characteristics, and types of treatment were compared between the group of patients with distant metastases and the comparative group using the Mann-Whitney U test for continuous and ordinal variables and by Pearson’s chi-squared test for categorical variables. The Shapiro-Wilk W test was used to determine whether HAX1 protein levels measured by immunohistochemistry were normally distributed. The associations between HAX1 immunoreactivity and progression, along with clinicopathological parameters, ER, PGR, and HER2 status, were assessed using the Mann-Whitney U test or Kruskal-Wallis test, depending on whether the nominal variable had two or more categories. If significant, the Kruskal-Wallis test was followed by pairwise comparisons using the Mann-Whitney U test. Receiver operating characteristic (ROC) curve analysis was performed to determine the overall test performance and to calculate possible cutoff points for HAX1 protein levels. Optimal cutoff values were calculated using the nearest to (0,1) method and the maximum value of the Youden index. Kaplan-Meier survival analyses were carried out for overall survival (OS) and distant relapse-free survival (DRFS). The log-rank test was used to evaluate the equality of survivor function for groups with lower and higher HAX1 expression categorized according to values obtained in the ROC curve analysis. The Cox proportional hazards model was used for univariate and multivariate analyses of patient survival depending on HAX1 expression, categorized as described above. In the multivariate survival analyses, HAX1 levels were assessed along with the following variables: PGR expression (categorization: positive vs. negative), clinical stage (I vs. II vs. III), histological grade (1 vs. 2 vs. 3), and molecular subtype (luminal vs. others). Hazard ratios (HR) with 95% confidence intervals and p-values were reported (Table 1). HAX1 gene expression retrieved from databases was compared in different subgroups using Student’s t-test and one-way Welch’s or Fisher’s ANOVA followed by post-hoc Tukey-Kramer test. All tests used in this study were two-tailed and the significance level (alpha) was always set to 0.05. Copyright © 2019 Alicja Trebinska-Stryjewska et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
1687-8450
eISSN
1687-8469
DOI
10.1155/2019/6375025
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Oncology Volume 2019, Article ID 6375025, 13 pages https://doi.org/10.1155/2019/6375025 Research Article Cytoplasmic HAX1 Is an Independent Risk Factor for Breast Cancer Metastasis Alicja Trebinska-Stryjewska, Lukasz Szafron, Alina Rembiszewska , Maciej Wakula ,SylwiaTabor , Renata Sienkiewicz, Joanna Owczarek , Anna Balcerak , Anna Felisiak-Golabek ,and EwaA.Grzybowska Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Roentgena , - Warsaw, Poland Correspondence should be addressed to Ewa A. Grzybowska; ewag@coi.waw.pl Received 31 July 2018; Revised 20 February 2019; Accepted 7 March 2019; Published 10 April 2019 Guest Editor: Peramaiyan Rajendran Copyright © 2019 Alicja Trebinska-Stryjewska et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. HAX1 is an antiapoptotic factor involved in the regulation of cell migration and calcium homeostasis, overexpressed in several cancers, including breast cancer. It has been suggested that HAX1 is also implicated in metastasis. Herein we report the results of meta-analysis of HAX expression, based on publicly available data, which confirms its significant overexpression in breast cancer and demonstrates copy number gain and prognostic value of HAX overexpression for metastatic relapse in ER+ tumors. IHC analysis reported here also reveals its significant overexpre ssion in breast cancer samples from primary tumors, indicating significantly higher HAX1 protein levels in a group of patients who developed distant metastases in a disease course. Moreover, we demonstrate that HAX1 localization is important for the prediction of metastatic relapse and that cytoplasmic but not nuclear HAX1 is an independent risk factor for breast cancer metastasis. 1. Introduction dormant for as long as 20 years but eventually may start to proliferate. Late recurrences were observed in as much as 50% Breast cancer is the most common neoplasm and the primary of these cancers [4]. Basal cancers tend to metastasize early cause of cancer death in women [1]. Breast cancer mortality is (with a peak about 2-3 years aeft r diagnosis) and frequently almostexclusivelyduetometastaticdisease[2,3].Thecurrent [8, 9], but typically there is no recurrence aeft r 5 years in this 5-year survival for primary breast cancer is quite high (80- subtype. Distinct pattern of metastatic relapse in basal and 92%), but, despite the advances in diagnosis and treatment of luminal subtypes suggests different routes for metastasis. early breast cancer patients, about 20-40% experience distant Better molecular characteristic of the primary tumor organ metastases, for which the prognosis is signica fi ntly is crucial for a good prediction of the clinical outcome. worse [4–7]. Breast cancer is heterogeneous disease and Genetic tests such as MammaPrint (for luminal and basal a probability to develop metastases depends not only on cancers) [10] and Oncotype DX (luminal cancers only) [11] histopathological parameters (lymph node status, histologic were developed as a diagnostic tool to predict risk of breast grade, and tumor size) but also on molecular subtypes cancer metastasis, based on mRNA expression signature of defined roughly as basal-like, normal-like, HER2 −enriched, selected gene sets (70 and 21 genes, resp.). Quantification of and luminal A and luminal B, each of which has a different the risk of recurrence is especially important for selecting a prognosis and a pattern of recurrence. For luminal cancers subset of luminal patients who may benefit from additional (estrogen and/or progesterone receptor positive) the progno- chemotherapy andsparing thosewhowouldnot. sis is better due to a very eeff ctive adjuvant endocrine therapy Two other factors have both prognostic and predictive and the fact that the metastases appear late, oeft n many years values in breast cancer and are commonly used in risk- aer ft initial diagnosis. Disseminated tumor cells could stay assessment: urokinase plasminogen activator protein (uPA) 2 Journal of Oncology and its inhibitor (PAI-1). ELISA-based assay, developed to met the inclusion and exclusion criteria were selected: 23 assess the levels of both proteins in breast cancer tissues, consecutive patients with distant metastases and 23 consec- allows to stratify the patients with node-negative disease into utive patients without distant metastases within the follow- a low-risk group, with a good prognosis without adjuvant up period (comparative group). Assuming a power of 80%, chemotherapy and a high-risk group, with high expression of 𝛼=0.05, and normal distribution of data, a number of patients both markers, who may benetfi from chemotherapy [12, 13]. tested would allow to detect a difference of at least 0.43 Additionally, to monitor a response to the treatment and between group means (with a common standard deviation to assess the probability of metastasis, several blood-based equalto0.5)whenanalyzedwithMann-WhitneyUtest. biomarkers have been developed, including Human Epider- Patients’ age at the time of diagnosis, estrogen receptor mal Growth Factor Receptor 2 (HER2), Cancer Antigen 15- (ER) status, progesterone receptor (PGR) status, HER sta- 3 (CA 15-3, MUC1), and Carcinoembryonic Antigen (CEA) tus, tumor size (pTNM scale), node status (pTNM scale), [14]. clinical stage (AJCC Anatomic Stage Group), histological Herein we assessed the potential of HAX1 expression level grade (Nottingham Histologic Score system), histology, and in primary tumor samples as an independent prognostic fac- molecular subtype (based on routine immunohistochemical tor for breast cancer metastasis. HAX1 was rst fi characterized evaluation of ER, PGR, HER2 and Ki-67) were recorded. in 1997 as an antiapoptotic factor [15] and several reports Patients’ characteristics are shown in Table S1. confirm its involvement in the regulation of apoptosis [16– 18]. Additionally, HAX1 was implicated in the regulation of .. Follow-Up and Outcomes. Patients’ records were tracked cell motility [19–21] and calcium homeostasis [22]. from the time of surgery until May 2016. Information about HAX1 overexpression was reported in several cancers clinical outcomes (distant metastases confirmed by imaging [23–25], including breast cancer [26, 27], and its role in or histologic evidence, death from any cause) was retrieved metastasis was suggested in some reports [20, 28]. Sheng and from clinical records and The National Cancer Registry in Ni [28] reported that higher HAX1 expression was related to Poland. Distant relapse-free survival (DRFS) was defined a lower 10-year survival rate in breast cancer patients. according to STEEP system [34] as the time from surgical In this report we present data analysis which confirms intervention until the time of distant recurrence, death from significant HAX1 overexpression in breast cancer samples, any cause, or the last follow-up. Complete events in DRFS coinciding with high amplification of the HAX gene. More- analysis were distant metastasis or death, whichever came over, the analyses reveal significant difference between HAX1 rfi st. Overall survival (OS) was dene fi d as the time from levels in primary tumor samples between nonmetastatic surgery to the last follow-up (censored event) or to the time and metastatic groups of patients, indicating that HAX1 of death from any cause (complete event). may represent an independent risk factor for breast cancer metastasis. Additionally, it was demonstrated that IHC assay .. Immunohistochemistry. Immunohistochemical staining which takes into account protein localization may predict with a monoclonal mouse anti-HAX1 antibody (BD Bio- clinical outcome more precisely and with a higher strength sciences) or a control mouse IgG of the same subclass was than mRNA-based estimations. performed as described previously [27] on a set of represen- tative slides from formalin-fixed, paraffin-embedded breast 2. Materials and Methods tumors. HAX-1 expression was scored manually according to Ball et al. [35]. It was evaluated independently for nuclear and .. Study Group. Formalin-fixed, paraffin-embedded (FFPE) cytoplasmic staining. Light microscopy evaluation at 400x tissue samples were collected from breast cancer patients magnicfi ation was used to count 100 tumor cells within areas receivingsurgicalinterventionattheMariaSklodowska- of the strongest staining. Each nucleus and cytoplasm in a CurieInstitute,OncologyCenter, betweenJanuary2007 and given eld fi was assigned to an intensity category of 0 (absent), May 2007, aer ft informed consent. The study was approved 1 (weak), 2 (moderate), or 3 (strong). The percentage of cells by Ethics Committee from the Maria Sklodowska-Curie in each intensity category was determined as N0, N1, N2, Institute, Oncology Center, Warsaw, in accordance with the and N3, respectively. A distribution score (ID score) was then guidelines of the Helsinki Declaration of 1975, revised in calculated as 1983. Clinical data and histology reports for each patient were reviewed by two clinicians and a pathologist, respectively. (/N0∗0/+/N1∗1/+/N2∗2/+/N3∗3/) (1) ID= . De-identified patients data were accessed using MedStream Designer platform (Transition Technologies S.A.). Patients eTh ID score therefore ranged from 0 (absent staining in all were retrospectively analyzed and divided into two groups: cells) to a maximum 3 (100% cells having a staining intensity with and without distant metastases within the follow-up of 3). eTh values of total HAX1 staining were obtained by period of 9 years. Inclusion criteria for the study were as adding nuclear and cytoplasmic ID scores for each sample. follows: invasive breast cancer stages I-III, the absence of distant metastasis at the time of surgery, and the presence of tumor tissue in FFPE confirmed by a pathologist. Exclusion . . Immunofluorescence. MCF7 (ATCC), MDA-MB-231 criteria were as follows: previous history of breast cancer, (ATCC), and HeLa (ATCC) human cell lines were used in previous or simultaneous history of any other malignances, the experiments. All cell lines were authenticated by Eurofins and neoadjuvant chemotherapy. A total of 46 patients who Genomics (Germany). Cells were grown in Dulbecco’s Journal of Oncology 3 Modified Eagle Medium supplemented with 10% fetal bovine Survival analysis of patients with metastatic relapse serum (Thermo Fisher Scientific). Immunouo fl rescence information was also performed using Breast Cancer Gene- was performed as described previously [36] with primary Expression Miner v4.1 (bcGenExMiner v4.1) [33]. Patients anti-HAX1 antibody (rabbit, 1:100, eTh rmo Fisher Scienticfi ) were split into two groups according to gene’s expression and secondary goat anti-rabbit Alexa Fluor 594 antibody median and Kaplan-Meier survival curves were plotted for (1:500, er Th mo Fisher Scientific). Cells were observed using each group. Breast Cancer Gene-Expression Miner v4.1 was the Zeiss LSM 800 confocal microscope. Images represent also used in targeted prognostic analysis of HAX gene single Z-stacks. Colocalization was quantified using ImageJ expression for all patients with metastatic relapse informa- JACoP plugin [37], for 3-5 independent elds fi of vision and tion. eTh results summarize univariate Cox scores (hazard approximately 25-60 cells per efi ld. ratios, p-values) for each cohort fulfilling the chosen criteria and all of these cohorts pooled together. eTh results were .. Database Gene Expression Analysis. HAX gene expres- presented in forest plot. Additionally, multivariate Cox scores sion in primary breast cancer compared to normal breast (adjusted on NPI/AOL) were calculated for HAX. tissue was analyzed using the Oncomine Platform (eTh rmo Fisher, Ann Arbor, MI) [38, 39]. HAX expression (RNAseq) .. Statistical Analysis. Statistical analysis was performed in breast cancer in relation to phenotypic variables was using SAS Enterprise Guide 7.11 (Copyright ©2015 by SAS explored in TCGA-BRCA cohort (data generated by the Institute Inc., Cary, NC, USA) and Stata software (Stata- TCGA Research Network: http://cancergenome.nih.gov/) Corp., College Station, TX, StataCorp LP.). GraphPad Prism using Xena Functional Genomics Explorer (https://xena- version 6.07 for Windows (GraphPad Software, La Jolla, browser.net/, 2018) and in a set of microarray data using California USA, www.graphpad.com) was used to visual- bcGenExMiner (http://bcgenex.centregauducheau.fr, 2018) ize data. O’Brien-Castelloe approximation (SAS Enterprise [33]. Guide7.11) wasemployedfor statisticalpower andsamplesize analysis for Mann-Whitney U test. Baseline demographics, . . Database Copy Number Variation and Mutation Analysis. tumor characteristics, and types of treatment were compared Genomic alterations (mutations, gene amplification, and/or between the group of patients with distant metastases and the deletion) of HAX gene in breast cancer were assessed using comparative group using the Mann-Whitney U test for con- cBioPortal for Cancer Genomics (http://www.cbioportal.org/ tinuous and ordinal variables and by Pearson’s chi-squared index.do, 2018) [40, 41]. eTh following cohorts of invasive test for categorical variables. The Shapiro-Wilk W test was breast carcinoma were included: Breast Cancer (METABRIC) used to determine whether HAX1 protein levels measured by [29, 42], Breast Invasive Carcinoma (British Columbia) [43], immunohistochemistry were normally distributed. The asso- Breast Invasive Carcinoma (Broad) [44], Breast Invasive ciations between HAX1 immunoreactivity and progression, Carcinoma (Sanger) [45], Breast Invasive Carcinoma (TCGA, along with clinicopathological parameters, ER, PGR, and PanCancer Atlas) [30], Mutational profiles of metastatic HER2status,wereassessedusingtheMann-WhitneyUtestor breast cancer (France) [46], and the Metastatic Breast Cancer Kruskal-Wallis test, depending on whether the nominal vari- Project (Provisional, April 2018). eTh other cohorts were able had two or more categories. If significant, the Kruskal- excluded from the analysis due to patients overlapping Wallis test was followed by pairwise comparisons using or the difference in sample type (xenografts instead of the Mann-Whitney U test. Receiver operating characteristic primary tumor). Groups with shallow deletion (possibly (ROC) curve analysis was performed to determine the overall heterozygous deletion), diploid status, gain, or high-level test performance and to calculate possible cutoff points for amplicfi ation of HAX gene generated by GISTIC algorithm HAX1 protein levels. Optimal cutoff values were calculated [47] were compared for mRNA expression in METABRIC using the nearest to (0,1) method and the maximum value and TCGA cohorts. For the latter, HAX mRNA levels were of the Youden index. Kaplan-Meier survival analyses were also correlated with log2 copy number values using Pearson’s carried outfor overallsurvival(OS)and distantrelapse- correlation coefficient. free survival (DRFS). eTh log-rank test was used to evaluate HAX copy number variation in primary breast cancer theequalityofsurvivorfunctionforgroups with lowerand in comparison to normal tissues was also analyzed on the higher HAX1 expression categorized according to values Oncomine Platform using the following threshold values: p- obtained in the ROC curve analysis. The Cox proportional value 0.05, fold change ‘all’, and gene rank ‘top 5%.’ hazards model was used for univariate and multivariate analyses of patient survival depending on HAX1 expression, .. Database Survival and Prognostic Analysis. Survival anal- categorized as described above. In the multivariate survival yses of patients stratified according to HAX expression were analyses, HAX1 levels were assessed along with the following performed using KM Plotter (http://kmplot.com/analysis/, variables: PGR expression (categorization: positive vs. nega- 2018) [48]. HAX1 expression levels based on Affymetrix probe tive), clinical stage (I vs. II vs. III), histological grade (1 vs. ID 201145 at in35 cohortsofbreastcancerpatientsdeposited 2 vs. 3), and molecular subtype (luminal vs. others). Hazard in GEO database (Gene Expression Omnibus, NCBI) were ratios (HR) with 95% confidence intervals and p-values used. eTh general settings were as follows: patients split by were reported (Table 1). HAX gene expression retrieved median or by best cut-off; follow-up threshold: all; probe set from databases was compared in different subgroups using options: only JetSet best probe set; quality control: removing Student’s t-test and one-way Welch’s or Fisher’s ANOVA redundant samples and excluding biased arrays. followed by post-hoc Tukey-Kramer test. All tests used in this 4 Journal of Oncology Table 1: Multivariate Cox regression analysis of HAX1 levels in human breast cancers. Evaluation of HAX levels in the cell nuclei Variable name OS (/ ) DRFS (/ ) HR (95% CI), p HR (95% CI), p HAX1 (≤1.05 vs.>1.05) NS NS Histological grade (2 vs. 1) 6.06E+8 (2.26E+8-1.63E+9),<0.001 NS Clinical stage (III vs. I) 3.171 (1.017-9.884), 0.047 NS Evaluation of HAX levels in the cytoplasm Variable name OS (/ ) DRFS (/ ) HR (95% CI), p HR (95% CI), p HAX1 (≤1.02 vs.>1.02) NS . (.- . ), . Histological grade (2 vs. 1) 4.86E+8 (1.74E+8-1.36E+9),<0.001 NS Evaluation of HAX levels in both of the nuclei and the cytoplasm Variable name OS (/ ) DRFS (/ ) HR (95% CI), p HR (95% CI), p HAX1 (≤2.06 vs.>2.06) NS . (.  -. ), . Histological grade (2 vs. 1) 3.44E+8 (1.17E+8-1.01E+9),<0.001 NS eTh multivariate analysis of prognosis was carried out using the Cox proportional hazards model. Values before and after a slash (/) stand for the number of complete observations versus all observations, respectively. Only the results with p-values<0.05 are shown and those with p-values<0.05 for HAX1 expression are highlighted in italic type. HR and CI stand for the hazard ratio and confidence interval, respectively. OS: overall survival; DRFS: distant recurr ence-free survival; NS: a nonsignificant result (p ≥0.05). study were two-tailed and the significance level (alpha) was mutations in HAX sequence, one truncating E59X and two always set to 0.05. missense mutations, E39K and P259A. Majority of the iden- tified alterations comprised of high-level gene amplification which was detected in all 4 cohorts containing copy number 3. Results variationdata[29,30, 42,46] andrangedfrom5.16% to .. HAX Is Significantly Overexpressed in the Majority of 21.06% of all cases (average 16.01%) (Figure 2(a)). Addi- Analyzed Datasets of Breast Cancer Primary Tumor Samples. tionally, low-level HAX gene gain was identified in 36.81% HAX overexpression in primary breast cancer in comparison to 63.86% cases (average 45.92%) whereas shallow deletion to normal breast tissues was identified at mRNA [26, 27] and (possibly heterozygous deletion) was present in only 0% to protein level [27]. To further confirm this observation we per- 3.76% of patients (average 1.49%). Additionally, log-2 HAX formed analysis on breast cancer cohorts using the Oncomine gene copy number units were compared between blood, Platform [38, 39] and taking into account invasive breast breast, and invasive ductal and invasive lobular carcinoma in cancer samples (ductal and lobular). HAX gene expression in TCGA-BRCA cohort using the Oncomine Platform and were invasive primary tumor was significantly elevated compared foundtobeelevatedfor bothinvasive ductalcarcinoma(fold to normal tissue in 16 out of 19 analyses (Figure 1(a), legend in change: 1.273, p=9.26E-135, gene rank: top 1%, Figure 2(b)) Figure S1). Detailed analyses for ductal and lobular carcinoma and invasive lobular carcinoma (fold change: 1.297, p=1.13E- in selected datasets confirmed these conclusions (Figures 22, gene rank: top 2%, Figure 2(c)). 1(b)–1(e)). Two cohorts analyzed using cBioPortal, METABRIC [29], HAX expressioninbreastcancerinrelationtopheno- and TCGA-BRCA [30] contained gene expression informa- typicvariableswasassessedinasetofmicroarraydatausing tion so it was possible to relate HAX gene copy number Breast Cancer Gene-Expression Miner v4.1 (bcGenExMiner) with HAX mRNA level. In both cohorts mRNA expression [33]. This analysis revealed that HAX expression correlates differed significantly between putative groups with diploid positively with grade (Figure 1(f)), confirming our previous DNA content and HAX gene gain or amplification (Figures results, obtained on a small group of patients [27]. It was 2(d) and 2(e)). In TCGA cohort HAX1 log2 copy number also observed that HAX expression differs significantly values showed a moderate positive correlation with mRNA inmolecularsubtypesofbreastcancer, withthehighest expression (Pearson’s r=0.656, p<0.0001, Figure 2(f)). expression in basal and luminal B subtypes, associated with more aggressive neoplasm (Figure 1(g)). .. HAX Overexpression Is Associated with Cancer Relapse and Has Prognostic Impact on ER+ Subset. Survival analyses .. HAX Gene Copy Number Is Altered in Breast Can- of breast cancer patients stratified according to HAX expres- cer Patients. Analysis of HAX gene in 7 cohorts of inva- sion were performed using KM Plotter and microarray data sive breast carcinoma patients using cBioPortal for Cancer from 35 breast cancer cohorts from GEO (Gene Expression Genomics [40, 41] revealed HAX altered status in 15% Omnibus, NCBI). RFS (relapse-free survival) analysis includ- (549/3655) of sequenced cases. Only three patients had ing 3,951 patients showed a statistically significant difference Journal of Oncology 5 Comparison of HAX1 Across 19 Analyses Overexpression Median Rank p-Value Gene 1 5 10 25 25 10 5 1 1404.5 6.82E−4 HAX1 Not measured 12345672 8 9 10 11 12 13 14 15 16 17 18 190 (a) HAX1 expression in Zhao Breast HAX1 expression in TCGA Breast HAX1 expression in TCGA Breast (fold change 2.508, p=8.70E-4) (fold change 1.319, p=3.01E-17) (fold change 1.372, p=1.79E-9) 1.5 1.5 1.0 1.0 1.0 0.5 0.5 0.5 0.0 0.0 0.0 −0.5 −0.5 −0.5 −1.0 −1.0 −1.0 −1.5 −2.0 −1.5 −1.5 12 12 breast invasive ductal breast invasive ductal breast invasive lobular (n=3) (n=61) (n=61) breast carcinoma breast carcinoma breast carcinoma (n=38) (n=389) (n=36) (b) (c) HAX1 expression in Gluck Breast HAX1 expression in Curtis Breast HAX1 expression in Curtis Breast (fold change 1.349, p=0.005) (fold change 1.309, p=4.08E-35) (fold change 1.283, p=1.38E-17) 1.0 3.5 3.0 3.0 2.5 0.5 2.5 2.0 2.0 0.0 1.5 1.5 −0.5 1.0 1.0 0.5 −1.0 0.5 0.0 −1.5 −0.5 0.0 12 12 breast invasive breast invasive ductal breast invasive lobular (n=4) breast carcinoma (n=144) breast carcinoma (n=144) breast carcinoma (n=154) (n=1556) (n=148) (d) (e) Box plot of HAX1 expression Box plot of HAX1 expression according to according to SBR PAM50 subtypes TCGA (HAX1 expression) <0.0001 <0.0009 −2 −2 −4 −4 F=11.06 −6 −6 Ｊ<0.0001 Ｊ = 0.0047 Ｊ < 0.0001 SBR1 SBR2 SBR3 (No:) (546) (1431) (1316) (No:) (1060) (780) (1468) (1013) (681) PAM50 subtype (f) (g) Figure 1: HAX1 is overexpressed in primary breast tumor in comparison to normal breast tissue. (a-e) HAX expression in invasive breast cancer (ductal and lobular) in comparison to normal breast tissue assessed in publicly available datasets on the Oncomine Platform. (a) Comparison of HAX overexpression across 19 analyses. Dataset legend in Figure S1. (b-e) HAX overexpression in selected datasets [29– 32]. Differences between groups were assessed by Student’s t-test and results with p-values <0.05 were considered significant. (f) HAX expression in breast cancer samples stratified according to grade (Scarff-Bloom-Richardson grade, SBR) analyzed using bcGenExMiner. (g) HAX expression in breast cancer samples stratified according to molecular subtype (PAM50 classification) in a set of microarray data analyzed using bcGenExMiner [33] (left panel) or RNAseq TCGA-BRCA data (right panel). Differences between groups in (f) and (g) were assessed by Welch’s or Fisher’s ANOVA followed by post-hoc Tukey-Kramer test. Basal-like (n=98) HER2-enriched (n=58) Luminal A (n=231) Luminal B (n=127) Normal-like (n=8) log2 median-centered ratio HAX1 mRNA level log2 median-centered ratio HAX1 mRNA level log2 median-centered intensity log2 median-centered ratio Basal-like HER2-E Luminal A Luminal B Normal breast-like log2 median-centered ratio HAX1, gene expression RNAseq log2 median-centered intensity Illumina Hiseq [unit: log2(norm_count+1) 6 Journal of Oncology 20% 15% 10% HAX1 copy number in TCGA Breast HAX1 copy number in TCGA Breast invasive ductal breast carcinoma invasive lobular breast carcinoma 5% 2.0 1.0 1.5 0.5 1.0 0.5 0.0 0.0 −0.5 −0.5 1 23 123 blood breast invasive ductal blood breast invasive lobular Mutation breast carcinoma (n=702) (n=111) (n=702) (n=111) breast carcinoma Amplification (n=693) (n=71) (a) (b) (c) METABRIC (HAX1 expression) TCGA (HAX1 expression) <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 6 14 <0.0001 <0.0001 TCGA (CNV vs mRNA) <0.0001 12 −2 −2 R=0.656 (95%CI, 0.620-0.689) F=121.30 F=230,54 −2 Ｊ<0.0001 −4 Ｊ<0.0001 Ｊ<0.0001 −4 −6 −4 −1 0 1 2 3 4 HAX1, log2 copy-number values HAX1, Putative copy-number alterations HAX1, Putative copy-number alterations (e) (f) (d) Figure 2: HAX gene copy number is altered in breast cancer patients. (a) Alterations in HAX gene analyzed in 7 cohorts of invasive breast carcinoma patients using cBioPortal for Cancer Genomics. (b-c) HAX gene copy number in TCGA-BRCA data from the Oncomine Platform for (b) invasive ductal carcinoma and (c) invasive lobular carcinoma compared to blood and normal breast tissue. (d-e) Comparison of HAX expression in primary breast cancer samples in relation to DNA copy number in (d) METABRIC cohort [29] and (e) TCGA-BRCA cohort [30]. Differences between groups were assessed by Fisher’s ANOVA followed by post-hoc Tukey-Kramer test. (f) Correlation of HAX expression and log2 copy number values in TCGA-BRCA cohort (Pearson’s correlation coefficient). in survival, favoring patients with lower HAX expression significance was detected for ER+ (n=2,061) but not ER- regardless of whether patients were split by median (HR=1.37, (n=801) subgroup of patients (HR=1.18, 95% CI, 1.00-1.39, 95% CI, 1.22-1.52, log-rank p=2.2E-08, Figure 3(a)) or best log-rank p=0.044 and HR=1.12, 95% CI, 0.89-1.40, log-rank cutoff value (HR=1.42, 95% CI, 1.27-1.58, log-rank p=3.6E-10, p=0.33, resp., Figure 3(a)). FDR=1%). OS (overall survival) analysis in 1,402 patients also Prognostic analysis was carried out using bcGenExMiner indicated statistically significant more favorable prognosis for [33]. Targeted prognostic analysis for HAX in a group patients with lower HAX expression, but only if patients of ER-positive patients with metastatic relapse information were split by best cutoff value (HR=1.41, 95%CI, 1.12-1.77, log- (n = 2,822) revealed statistical significance (HR=1.15, p- rank p=0.0034) and at the expense of false discovery rate value=0.0008) for HAX expression level in the pooled (FDR=50%) (Figure S2A). cohort (Figure 3(b), left panel). Additionally, to evaluate In breast cancer ER status is one of the most important independent prognostic impact of HAX in ER+ patients prognostic and predictive factors. er Th efore, RFS analysis relative to the well-established breast cancer prognostic wasperformed usingKMPlotteronsubgroupsofbreast indexes, including Nottingham Prognostic Index (NPI) [49] cancer patients with different ER status, using median value and Adjuvant! Online (AOL) [50], adjusted Cox proportional of HAX expression to avoid high values of FDR. Statistical hazards model was used, revealing statistical significance for Shallow Deletion (n=25) Diploid (n=777) Gain (n=701) Amplification (n=401) Shallow Deletion (n=17) Diploid (n=273) Gain (n=686) Amplification (n=92) HAX1, mRNA Expression Z-Scores (Illumina Human v3 microarray) Alteration Frequency Breast (METABRIC) Breast (TCGAPanCan) e MBC Project BRCA (INSERM2016) Breast (BCCRC 2012) Breast (Broad 2012) Breast (Sanger) log2 copy number units HAX1, mRNA Expression Z-scores (Illumina HiSeq_RNASeqV2 syn4976369) log2 copy number units HAX1, mRNA Expression Z-Scores (Illumina HiSeq_RNASeqV2 syn4976369) Journal of Oncology 7 Total ER+ ER- HR = 1.12 (0.89 - 1.4) HR = 1.37 (1.22 − 1.52) HR = 1.18 (1 - 1.39) log-rank P = 0.33 log-rank P = 0.044 log-rank P = 2.2e−08 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 Time (months) Time (months) Time (months) Expression Expression Expression low low low high high high (a) ER+ ER- (b) Figure 3: HAX overexpression is higher in patients with breast cancer relapse and has prognostic impact on ER+ subset of patients. (a) RFS analysis for total number of patients and subsets with different ER status (left: ER+; right: ER-). Patients were split into groups with high and low HAX expression (based on microarray data, split by median). Kaplan-Meier estimates were generated in KM Plotter online tool for all data available for 2017 (merged datasets). Probability of cancer relapse is plotted against time. (b) Forest plots estimating prognostic impact of HAX expression in ER+ (n = 2,822) and ER- (n = 1,072) subsets of patients with metastatic relapse information (bcGenExMiner). Values in columns represent summarized univariate Cox scores (p-values, hazard ratios) for each cohort fulfilling the chosen criteria and for pooled cohorts. MR: metastatic relapse. HAX expressionadjustedonAOL(HR1.27, 95%CI1.06-1.52, were consistent with the previous RFS analyses, showing p-value: 0.0108, 12 cohorts, 382 patients, 101 metastases). signicfi ant difference for ER+ group of patients and the lack The same analysis performed for patients with negative of significance in ER- group of patients (Figure S2C). ER status (n=1,072) revealed the lack of statistical signif- icance (p-value=0.3853 for the pooled cohort) and even . . Cytoplasmic HAX Levels Are Significantly Higher in the the tendency for better prognosis associated with HAX1 Primary Tumor of Breast Cancer Patients Who Experience overexpression (Figure 3(b), right panel). Distant Metastasis during the Disease Course. 46 breast can- Prognostic analysis performed for HAX expression cer patients who were free of distant metastasis at the time regardless of ER status (n=3,924) indicated significance, but of surgery and received no neoadjuvant therapy were retro- bordering on the 0.05 threshold (HR=1.07, 95% CI, 1.00- spectively analyzed for HAX1 protein levels (cytoplasmic and 1.14, p-value: 0.0432, Figure S2B). Additionally, KM curves nuclear) in primary tumors by immunohistochemistry. Half for metastatic relapse-free survival (MRFS) were plotted in of the analyzed group developed distant metastasis during a bcGenExMiner for each group of patients with metastatic follow-up period of 9 years. Cytoplasmic HAX1 protein levels relapse information (all patients, ER+, ER-), and the results were significantly elevated (p=0.0003) in the group of patients Probability 0.0 0.2 0.4 0.6 0.8 1.0 Probability 0.0 0.2 0.4 0.6 0.8 1.0 Probability 0.0 0.2 0.4 0.6 0.8 1.0 8 Journal of Oncology p=0.0003 p=0.0761 p=0.0093 4 4 3 3 2 2 1 1 0 0 −1 −1 −1 distant distant distant distant distant distant metastasis-free metastasis metastasis-free metastasis metastasis-free metastasis (n=23) (n=23) (n=23) (n=23) (n=23) (n=23) (a) (b) (c) p=0.0166 distant metastasis distant metastasis-free p=0.0154 p=0.1335 p=0.0844 p=0.0379 p=0.0626 HAX1 IgG −1 −1 grade 1 grade 2 grade 3 grade 1 grade 2 grade 3 (n=5) (n=23) (n=16) (n=5) (n=23) (n=16) (d) (e) (f) Figure 4: HAX1 protein level in primary tumors stratified according to selected clinical and histological factors (presence of distant metastases, tumor grade). (a-d) HAX1 protein levels in the primary tumor were quantified from IHC data in distant metastasis-free versus distant metastasis group for (a) cytoplasmic, (b) total, and (c) nuclear HAX1 staining. (d) Representative images of HAX1 IHC and negative isotype control for patients from metastasis-free versus distant metastasis group.×40 objective, bar: 100𝜇m. (e) Cytoplasmic and (f) total HAX1 staining in breast cancers stratified according to tumor grade (grades 1-3). Results for individual patients and median and interquartile range for each group are shown. Differences in HAX1 protein levels between groups were assessed by the Mann-Whitney U test and results with p-values<0.05 were considered significant. with distant metastasis (median 1.50, mean±SD 1.48±0.92, .. High Cytoplasmic and Total HAX Protein Levels in Breast 95% CI of the mean 1.08-1.87) compared to the group with no Cancer Cells Are Risk Factors for Distant Metastasis and Death. distant metastasis (median 0.40, mean±SD 0.50±0.54, 95% To ascertain if HAX1 protein levels in primary tumor can CI of the mean 0.26-0.73) (Figure 4(a)). Total HAX1 staining be used as a prognostic factor in breast cancer, we analyzed wasalsohigherinthe distantmetastasispatient group, follow-up patient data and recorded time to distant recur- although the eeff ct was less prominent (metastasis: median rence and/or time to death from any cause for all 46 patients. 1.94, mean±SD 1.93±0.85, 95% CI of the mean 1.56-2.29 vs. The total follow-up time was 9 years; 61% of the patients metastasis-free: median 1.37, mean±SD 1.22±0.76, 95% CI of had been followed for a minimum of 5 years. 23 patients the mean 0.89-1.55, p=0.0093) (Figure 4(b)). eTh opposite developed distant metastasis. 23 out of 46 patients were still effect, albeit not statistically significant, was observed for aliveattheendofthefollow-upperiod(18 in agroup with no nuclear HAX1 levels (metastasis: median 0.00, mean±SD distant metastasis and 5 in a group with distant metastasis). 0.45±0.60, 95% CI of the mean 0.19-0.71 vs. metastasis-free: Receiver operating characteristic (ROC) analysis was median 1.00, mean±SD 0.73±0.66, 95% CI of the mean 0.44- performed to define the best cutoff value of the HAX1 signal 1.01, p=0.0761) (Figure 4(c)). Representative images of the andtomeasuretheoveralltest performancewhich would typical staining in metastasis-free and metastatic groups are use HAX1 protein levels to predict breast cancer metastasis. presented in Figure 4(d). eTh analysis was done separately for cytoplasmic, total, eTh two analyzed groups were well matched, as patients’ and nuclear HAX1 immunohistochemical staining (Figures clinicopathological parameters and treatment did not differ 5(a)–5(c)). The highest value of area under the curve (AUC) significantly except for PGR status (Table S1). Analyses of was obtained for cytoplasmic HAX1: 0.7977 (95% CI 0.6628- HAX1 protein levels in groups stratiefi d according to known 0.9327, p=0.0005) (Figure 5(a)). eTh best cutoff points for prognostic factors showed that the values of the cytoplasmic cytoplasmic, nuclear, and total HAX1 were, respectively, 1.02 and total HAX1 signal were positively associated with tumor (sensitivity 0.65 and specificity 0.87), 1.05 (sensitivity 0.82 grade (Figures 4(e) and 4(f), resp.), but not other prognostic and specificity 0.48), and 1.49 (sensitivity 0.78 and specificity factors. 0.57). HAX1 cytoplasmic staining HAX1 total staining HAX1 cytoplasmic staining HAX1 nuclear staining HAX1 total staining Journal of Oncology 9 ROC of HAX1 cytoplasmic staining ROC of HAX1 total staining ROC of HAX1 nuclear staining 1.0 1.0 1.0 0.5 0.5 0.5 AUC (95% CI)=0,7977 (0,6628-0,9327) AUC (95% CI)=0,7212 (0,5742-0,8678) AUC (95% CI)=0,6456 (0,4835-0,8076) p=0,0907 p=0,0005 p=0,0102 0.0 0.0 0.0 0.0 0.5 1.0 0.0 0.5 1.0 0.0 0.5 1.0 1 - Specificity 1 - Specificity 1 - Specificity Sensitivity Sensitivity Sensitivity identity identity identity (a) (b) (c) HAX1 cytoplasmic staining HAX1 nuclear staining 110 HAX1 total staining p=0.0012 p=0.0074 p=0.3400 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 DRFS (months) DRFS (months) DRFS (months) cytoplasmic HAX1 ID score ≤1.02 total HAX1 staining ≤1.49 HAX1 nuclear staining ≤1.05 cytoplasmic HAX1 ID score >1.02 total HAX1 staining >1.49 HAX1 nuclear staining >1.05 HAX1 nuclear staining HAX1 total staining 110 110 HAX1 cytoplasmic staining 100 100 90 90 80 80 70 70 60 60 30 30 20 20 20 p=0.4456 p=0.0097 p=0.0134 10 10 0 0 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 OS (months) OS (months) OS (months) HAX1 cytoplasmic staining ≤1.02 HAX1 nuclear staining ≤1.05 HAX1 total staining ≤1.49 HAX1 cytoplasmic staining >1.02 HAX1 nuclear staining >1.05 HAX1 total staining >1.49 (f) (d) (e) Figure 5: HAX1 protein level in primary tumor is a risk factor for breast cancer progression. (a-c) Receiver operating characteristic analysis for (a) cytoplasmic, (b) total, and (c) nuclear HAX1 protein levels was performed to define the best cutoff values for subsequent survival analysis. Area under curve (AUC) with 95% CI and p-values for each ROC curve are shown. (d-f) Kaplan-Meier survival estimates for distant recurrence-free survival (DRFS) and overall survival (OS) in breast cancer patients according to proposed cutoff values of (d) cytoplasmic HAX1 protein levels≤1.02 (n=28) versus>1.02 (n=18), (e) total HAX1 protein levels≤1.49 (n=18) versus>1.49 (n=28), and (f) nuclear HAX1 protein levels≤1.05 (n=31) versus>1.05 (n=15). eTh log-rank test was used to evaluate the equality of survivor function for groups with lower and higher HAX1 expression and p-values<0.05 were considered significant. Cutoff points estimated from ROC curves were used in compared to 89% of patients in the group with cytoplasmic subsequent survival analyses by the Kaplan-Meier method. HAX1 protein levels>1.02. Overall survival analysis showed The log-rank test showed a significant difference favoring, that,attheendoffollow-up,64% ofpatientswerestillalivein for both distant recurrence-free survival (DRFS) and overall the group with a cytoplasmic HAX1 of≤1.02 compared to 28% survival (OS), patients with a cytoplasmic HAX1 ID score of patients in the group with cytoplasmic HAX1>1.02. Similar of≤1.02 (p=0.0012 and p=0.0134, resp., Figure 5(d)). 43% of results were observed for total HAX1 protein levels. Patients patients in the group with cytoplasmic HAX1 protein levels with total HAX1 protein levels ≤1.49 showed significantly ≤1.02 experienced distant metastasis/death within 9 years increased DRFS and OS compared to the group with a total Percent survival Percent survival Sensitivity Percent survival Percent survival Sensitivity Sensitivity Percent survival Percent survival 10 Journal of Oncology HAX1 Nuclei (DAPI) Merged Figure 6: HAX1 localization in breast cancer cell lines of different characteristics. Endogenous staining of HAX1 (red) and nuclei (DAPI, blue) in MCF7, luminal-like epithelial cells and MDA-MB-231, basal-like cells aer ft epithelial-mesenchymal transition. Bar: 20 𝜇m. HAX1 protein level of>1.49 (p=0.0074 and p=0.0097, resp., analysis to assess its eeff ct on metastasis. Database analysis Figure 5(e)). Nuclear HAX1 staining showed no prognostic on large group of patients conrfi med HAX overexpression in value for neither DRFS nor OS (Figure 5(f)). breast cancer samples, which tallies with the previous study Overall survival (OS) and distant recurrence-free survival by Luoetal. [26].Theanalysisrevealedalsothecorrelationof (DRFS) for 46 breast cancer patients were also evaluated HAX1 overexpression with tumor grade, which is consistent by 3 different univariate and multivariate analyzes, in which with our previous [27] and current IHC results. Additionally, the HAX1 protein expression, either cytoplasmic, nuclear, or HAX overexpression was shown to correlate with gene cumulative, was assessed by IHC (Table 1). We found out amplicfi ation. Although there were several studies reporting that elevated HAX1 levels in the cytoplasm emerged as an HAX overexpression in different types of malignancies, we independent, negative prognostic factor, associated with an showed for the first time that high HAX mRNA levels in increased risk of distant metastasis (HR 2.832, 95% CI 1.207- cancer cells could be a consequence of gene amplicfi ation, at 6.644, p=0.017). Correspondingly, the results obtained for least in breast cancer. Detailed analysis of HAX expression cumulative expression of HAX1 also showed its adverse effect in molecular subtypes demonstrated that the highest overex- on DRFS (HR 4.249, 95% CI 1.404-12.86, p=0.010). HAX1 pression was observed in basal and luminal B subtypes, which nuclear expression had no impact on survival. are more aggressive. Database analysis of HAX expression in correlation to . . HAX Localization Varies among Breast Cancer Cell Lines. metastasis revealed its signicfi ant prognostic value for lumi- Endogenous HAX1 protein was detected by immunouo- fl nal (ER+) subset while for ER-, despite high overexpression rescence in luminal-like MCF7 and basal-like MDA-MB- in basal cancers, the expression level had no prognostic value. 231 cell lines, revealing significant differences. In MCF7 cells This apparent paradox can be resolved on the basis of cellular HAX1 staining was mostly cytoplasmic, while in MDA-MB- localization. 231 HAX1 was also detected in the nuclei (Figure 6). Nuclear Our previous IHC analysis [27] indicated two dieff rent colocalization was calculated using ImageJ JACoP, showing localizationsofHAX1proteininbreastcancertumorsamples: a significant shift of Pearson’s correlation coefficient (PCC) cytoplasmic and nuclear. Nuclear localization of HAX1 was and two Mander’s overlap coefficients (M1, M2) from 0.101 also reported in cell lines [36] and rat tests [51]. Different (PCC), 0.207 (M1), and 0.116 (M2) in MCF7 cells to 0.467 localization may translate into different functionality and (PCC), 0.377 (M1), and 0.592 (M2) in MDA-MB-231 cells, different impact on tumor progression, as in case of Aurora respectively (p-values for PCC, M1, and M2: 0.0105, 0.0328, A kinase, where nuclear protein acquires kinase-independent and 0.0181, resp.). transactivating function, which enhances breast cancer stem cell phenotype [52]. u Th s, in this report, we have analyzed 4. Discussion HAX1 protein levels in the primary tumor of breast cancer Advancing on our previous study [27], in which we demon- patients divided into metastatic and nonmetastatic groups. strated HAX overexpression in breast cancer and its differen- IHC analysis enabled us to differentiate between cytoplas- tial localization (cytoplasmic and nuclear), we expanded our mic and nuclear localization of HAX1. Overall, our results MDA-MB-231 MCF7 Journal of Oncology 11 demonstrated that HAX1 protein level is significantly higher ausefultoolforestimating theprobability ofluminalbreast in metastatic group of patients, but this effect can be observed cancer dissemination. only for evaluations concerning cytoplasmic and total HAX1, while for nuclear localization it does not exist and the trend Data Availability is even opposite (less HAX1 in metastatic group). Clearly, the results for total HAX1 levels are inu fl enced by the cytoplasmic eTh data used to supportthe nfi dingsofthisstudy are subset, for which the difference is huge. available from the corresponding author upon request. uTh s, ourevaluationofHAX1 proteinlevelsandlocal- ization in the samples from metastatic versus nonmetastatic Conflicts of Interest groups of patients indicates a positive relationship between HAX1 cytoplasmic expression and the occurrence of a The authors declare no conflicts of interest. secondary tumor at distant locations in the course of the disease (opposite to the relations observed for Aurora A). Funding High cytoplasmic HAX1 level is associated negatively with progression-free survival and overall survival. Similar results This work was supported by the Polish National Science Cen- were obtained for total HAX1; however, ROC curve analysis ter Grants Nos. 2011/01/B/NZ1/03674, 2014/14/M/NZ1/00437, indicated a higher ability to identify patients at risk of and 2016/21/B/NZ2/03473. MedStream Designer (Transition progression when using cytoplasmic, not total HAX1 levels. Technologies S.A.) purchase was n fi anced by ONKO.SYS Accordingly, the experimental immunouo fl rescence project (Grant No. POIG.02.03.00-14-084/13) from the Polish results showing that HAX1 localization is more cytoplasmic National Centre for Research and Development. in luminal-like than basal-like cell lines can explain the apparent difference in HAX1 prognostic value for ER+ and ER- subsets. It seems plausible that, in basal cells, Acknowledgments despite the high HAX1 expression, nuclear localization Wewouldliketothank MariaZwierko,PhD forproviding of HAX1 prevents its prometastatic action, by assuming patients’ data from eTh National Cancer Registry in Poland. different functionality or simply by sequestering cytoplasmic HAX1. Alternatively, it seems plausible that nuclear HAX1 can block and/or sequester in inactive complexes some Supplementary Materials nuclear factor(s), specific to luminal cancers, whose action is linked to metastasis, probably by the regulation of Table S1: clinical and pathological characteristics of breast cancer patients. Figure S1: dataset legend to Oncomine transcription. u Th s, nuclear HAX1 would have protective meta-analysis of HAX expression in breast cancer (ductal effect (restricted to luminal-like cells), which would not and lobular). Figure S2: HAX overexpression is associ- be present in cells with cytoplasmic HAX1. Of note, HAX1 ated with metastatic relapse and survival (additional data). was shown to bind directly estrogen receptor [53], which (Supplementary Materials) is a prometastatic factor. Estrogen responsiveness is one of the main factors differentiating between the two cell lines used in experiments. However, since luminal patients are References treated with anti-estrogen therapy, some other factors may be [1] F. Bray, J. Ferlay, I. Soerjomataram et al., “Global cancer statistics involved. Further research should explain specific molecular 2018: GLOBOCAN estimates of incidence and mortality world- mechanisms of function for cytoplasmic and nuclear HAX1, wide for 36 cancers in 185 countries,” CA: A Cancer Journal for which contribute to metastasis. Clinicians,vol.68,no.6,ArticleID30207593,pp.394–424,2018. [2] C. L. Chaer ff and R. A. Weinberg, “A perspective on cancer cell 5. Conclusions metastasis,” Science,vol.331,no. 6024,pp.1559–1564,2011. [3] F.Cardoso,A.Costa,L.Nortonetal.,“1stInternational Overall, presented results indicate HAX overexpression, consensus guidelines for advanced breast cancer (ABC 1),” copy number gain, and prognostic impact on metastatic Breast,vol.21,no.3,ArticleID 22425534,pp.242-52,2012. breast cancer. Moreover, cytoplasmic but not nuclear HAX1 [4] X.H.Zhang,M.Giuliano,M.V.Trivedi,R. Schi,a ff ndC. K. demonstrated to be an independent, negative prognostic Osborne, “Metastasis dormancy in estrogen receptor-positive factor for breast cancer metastasis. breast cancer,” Clinical Cancer Research,vol.19, no.23, Article Multigene tests like MammaPrint and Oncotype DX have ID 24298069, pp. 6389–6397, 2013. a proved prognostic value, but they rely on expression on [5] Z. H. Li, P. H. Hu, J. H. Tu et al., “Luminal B breast cancer: mRNA level, which does not always accurately correspond to patterns of recurrence and clinical outcome,” Oncotarget,vol. protein level and does not consider the factor of protein local- 7, no. 40, Article ID 27542253, pp. 65024–65033, 2016. ization. It is possible that for some markers establishing the [6] R. L. Siegel, K. D. Miller, and A. Jemal, “Cancer statistics, 2015,” level of localized protein provides more accurate prediction CA: A Cancer Journal for Clinicians,vol.65,no.1,ArticleID than mRNA profiling. Further analysis is required to conrm fi 25559415, pp. 5–29, 2015. its prognostic value in the clinic, but assessing HAX1 protein [7] J. Ferlay, I. Soerjomataram, R. Dikshit et al., “Cancer incidence levels and localization in primary tumor samples may become and mortality worldwide: sources, methods and major patterns 12 Journal of Oncology in GLOBOCAN 2012,” International Journal of Cancer,vol.136, [22] E.Vafiadaki,D.A.Arvanitis,S.N.Pagakisetal.,“ea Th nti- no. 5, Article ID 25220842, pp. E359–86, 2014. apoptotic protein HAX-1 interacts with SERCA2 and regulates Its protein levels to promote cell survival,” Molecular Biology of [8] F. Bertucci, P. Finetti, and D. Birnbaum, “Basal breast cancer: the Cell (MBoC), vol. 20, no. 1, Article ID 18971376, pp. 306–318, A complex and deadly molecular subtype,” Current Molecular Medicine, vol. 12, no. 1, Article ID 22082486, pp. 96–110, 2012. [9] H. Kennecke, R. Yerushalmi, R. Woods et al., “Metastatic behav- [23] X.Li,J.Jiang,R.Yangetal.,“Expression of HAX-1incolorectal ior of breast cancer subtypes,” Journal of Clinical Oncology,vol. cancer and its role in cancer cell growth,” Molecular Medicine 28, no. 20, Article ID 20498394, pp. 3271–3277, 2010. Reports,vol.12,no.3,ArticleID26062578,pp.4071–4078, 2015. [10] M. J. van de Vijver, Y. D. He, L. J. van’t Veer et al., “A gene- [24] X. Li, T. Li, B. You et al., “Expression and function of HAX-1 in expression signature as a predictor of survival in breast cancer,” uman cutaneous squamous cell carcinoma,” Journal of Cancer, e New England Journal of Medicine,vol.347,no.25,Article ID vol. 6, no. 4, Article ID 25767605, pp. 351–359, 2015. 12490681, pp. 1999–2009, 2002. [25] H. Wu, J. Chen, Q. Wang et al., “Abnormal expression of HAX1 [11] S. Paik, S. Shak, G. Tang et al., “A multigene assay to predict is associated with cellular proliferation and migration in human recurrence of tamoxifen-treated, node-negative breast cancer,” hypopharyngeal squamous cell carcinoma,” Molecular Medicine e New England Journal of Medicine, vol. 351, no. 27, Article ID Reports, vol. 16, no. 4, Article ID 28791389, pp. 4664–4670, 2017. 15591335, pp. 2817–2826, 2004. [26] J. Hu, Z. Li, X. Li et al., “hSav1 interacts with HAX1 and [12] V. R. Jacobs, R. E. Kates, E. Kantelhardt et al., “Health eco- attenuates its anti-apoptotic effects in MCF-7 breast cancer nomic impact of risk group selection according to ASCO- cells,” International Journal of Molecular Medicine,vol.28, no. recommended biomarkers uPA/PAI-1 in node-negative pri- 3, Article ID 21567072, pp. 349–355, 2011. mary breast cancer,” Breast Cancer Research and Treatment,vol. [27] A. Trebinska, A. Rembiszewska, K. Ciosek et al., “HAX-1 138, no. 3, pp. 839–850, 2013. overexpression, splicing and cellular localization in tumors,” [13] F.Janicke,A.Prechtl,C.Thomssenetal.,“Randomizedadjuvant BMC Cancer, vol. 10, no. 76, Article ID 20196840, 2010. chemotherapy trial in high-risk, lymph node-negative breast [28] C. Sheng and Q. Ni, “Expression of HAX1 and Ki-67 in breast cancer patients identified by urokinase-type plasminogen acti- cancer and its correlations with patient’s clinicopathological vator and plasminogen activator inhibitor type 1,” JNCI: Journal characteristics and prognosis,” International Journal of Clinical of the National Cancer Institute,vol.93,no.12,ArticleID and Experimental Medicine,vol.8,no. 11,ArticleID26885017, 11416112, pp. 913-20, 2001. pp. 4–10, 2090. [14] R. Molina,J.M.Auge, J. M. Escudero et al., “Evaluation of tumor markers (HER-2/neu oncoprotein, CEA, and CA 15.3) [29] C. Curtis, S. P. Shah, S. F. Chin et al., “eTh genomic and tran- in patients with locoregional breast cancer: prognostic value,” scriptomic architecture of 2, 000 breast tumours reveals novel Tumor Biology,vol.31,no.3,ArticleID 20361287,pp.171–180, subgroups,” Nature,vol.486,no. 7403,Article ID 22522925,pp. 2010. 346–352, 2012. [15] Y. Suzuki, C. Demoliere, D. Kitamura et al., “HAX-1, a novel [30] N. Cancer Genome Atlas, “Comprehensive molecular portraits intracellular protein, localized on mitochondria, directly asso- of human breast tumours,” Nature,vol.490,no. 7418,Article ID ciates with HS1, a substrate of Src family tyrosine kinases,” e 23000897, pp. 61–70, 2012. Journal of Immunology, vol. 158, no. 6, Article ID 9058808, pp. [31] S. Gluck, J. S. Ross, M. Royce et al., “TP53 genomics predict 2736-44, 1997. higher clinical and pathologic tumor response in operable [16] Y. Han, Y. Chen, Z. Liu et al., “Overexpression of HAX-1 protects early-stage breast cancer treated with docetaxel-capecitabine cardiac myocytes from apoptosis through caspase-9 inhibition,” +/- trastuzumab,” Breast Cancer Research and Treatment,vol. Circulation Research,vol.99,no.4,ArticleID16857965,pp.415– 132, Article ID 21373875, pp. 781–791, 2012. 423, 2006. [32] H. Zhao, A. Langerod, Y. Ji et al., “Different gene expression [17] L. Cilenti, M. M. Soundarapandian, G. A. Kyriazis et al., patterns in invasive lobular and ductal carcinomas of the breast,” “Regulation of HAX-1 anti-apoptotic protein by Omi/HtrA2 Molecular Biology of the Cell (MBoC),vol.15,no.6,Article ID protease during cell death,” eJournalofBiologicalChemistry, 15034139, pp. 2523–2536, 2004. vol.279,no.48, pp.50295–50301,2004. [33] P. Jez ´ eq ´ uel, M. Campone, W. Gouraud et al., “bc-GenExMiner: [18] T. V. Sharp, H. W. Wang, A. Koumi et al., “K15 protein of Kaposi’s an easy-to-use online platform for gene prognostic analyses in sarcoma-associated herpesvirus is latently expressed and binds breast cancer,” Breast Cancer Research and Treatment,vol.131, to HAX-1, a protein with antiapoptotic function,” Journal of no.3,Article ID21452023,pp.765–775, 2012. Virology,vol.76,no.2,ArticleID 11752170,pp. 802–816,2002. [34] C. A. Hudis, W. E. Barlow, J. P. Costantino et al., “Proposal [19] P. J. Cavnar, E. Berthier, D. J. Beebe, and A. Huttenlocher, “Hax1 for standardized definitions for efficacy end points in adjuvant regulates neutrophil adhesion and motility through RhoA,” e breast cancer trials: the STEEP system,” Journal of Clinical Journal of Cell Biology, vol. 193, no. 3, Article ID 21518791, pp. Oncology,vol.25,no.15, ArticleID17513820, pp.2127–2132, 465–473, 2011. [20] A. G. Ramsay, M. D. Keppler, M. Jazayeri et al., “HS1-associated [35] E. Ball, J. Bond, B. Franc et al., “An immunohistochemical study protein X-1 regulates carcinoma cell migration and invasion of p16(INK4a) expression in multistep thyroid tumourigenesis,” via clathrin-mediated endocytosis of integrin 𝛼v𝛽6,” Cancer European Journal of Cancer,vol.43, no.1,Article ID 17046239, Research, vol. 67, no. 11, Article ID 17545607, pp. 5275–5284, pp.194–201,2007. [21] V. Radhika, D. Onesime, J. H. Ha, and N. Dhanasekaran, “G𝛼13 ´ [36] E. A. Grzybowska, V. Zayat, R. Konopinski et al., “HAX-1 is stimulates cell migration through cortactin-interacting protein a nucleocytoplasmic shuttling protein with a possible role in Hax-1,” eJournalofBiologicalChemistry,vol.279,no.47, mRNA processing,” FEBS Journal,vol.280,no. 1, ArticleID Article ID 15339924, pp. 49406–49413, 2004. 23164465, pp. 256–272, 2013. Journal of Oncology 13 [37] S. Bolte and F. P. Cordelier ` es, “A guided tour into subcel- [53] M.P.Walker,M.Zhang,T.P.Le, P. Wu,M.Laine, ´ and G. lular colocalization analysis in light microscopy,” Journal of L. Greene, “RAC3 is a pro-migratory co-activator of ER𝛼,” Microscopy,vol.224,no. 3,pp.213–232,2006. Oncogene, vol. 30, no. 17, pp. 1984–1994, 2011. [38] D.R. Rhodes,J.Yu, K.Shankeretal.,“ONCOMINE, acancer microarray database and integrated data-mining platform,” Neoplasia,vol.6,no.1, ArticleID15068665,pp.1–6,2004. [39] D. R. Rhodes, S. Kalyana-Sundaram, V. Mahavisno et al., “Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles,” Neoplasia,vol.9,no. 2, Article ID 17356713, pp. 166–180, 2007. [40] J. Gao, B. A. Aksoy, U. Dogrusoz et al., “Integrative analysis of complex cancer genomics and clinical profiles using the cBio- Portal,” Science Signaling,vol.6,no.269,ArticleID23550210, [41] E.Cerami,J.Gao,U.Dogrusozetal.,“ec Th Biocancergenomics portal: an open platform for exploring multidimensional cancer genomics data,” Cancer Discovery,vol.2,no.5,ArticleID 22588877, pp. 401–404, 2012. [42] B. Pereira, S. F. Chin, O. M. Rueda et al., “The somatic mutation profiles of 2, 433 breast cancers refines their genomic and transcriptomic landscapes,” Nature Communications,ArticleID 27161491, 2016. [43] S. P. Shah, A. Roth, R. Goya et al., “eTh clonal and mutational evolution spectrum of primary triple-negative breast cancers,” Nature,vol.486,no. 7403,ArticleID 22495314,pp. 395–399, [44] S. Banerji, K. Cibulskis, C. Rangel-Escareno et al., “Sequence analysis of mutations and translocations across breast cancer subtypes,” Nature, vol. 486, no. 7403, Article ID 22722202, pp. 405–409, 2012. [45] P. J. Stephens, P. S. Tarpey, H. Davies et al., “eTh landscape of cancer genes and mutational processes in breast cancer,” Nature, vol. 486, no. 7403, Article ID 22722201, pp. 400–404, 2012. [46] C. Lefebvre, T. Bachelot, T. Filleron et al., “Mutational profile of metastatic breast cancers: a retrospective analysis,” PLoS Med, vol. 13, no. 2, Article ID e1002201, 2016. [47] R. Beroukhim, G. Getz, L. Nghiemphu et al., “Assessing the significance of chromosomal aberrations in cancer: methodol- ogy and application to glioma,” in Proceedings of the National Academy of Sciences of USA, vol. 104, pp. 20007-12, 2007 Dec 11. [48] B. Gyorffy, A. Lanczky, A. C. Eklund et al., “An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients,” Breast Cancer Research and Treatment,vol.123,no.3, ArticleID 20020197, pp. 725–731, 2010. [49] M.H.Galea, R.W.Blamey,C.E.Elston,andI.O.Ellis,“eTh Nottingham prognostic index in primary breast cancer,” Breast Cancer Research and Treatment,vol.22,no.3,pp.207–219, 1992. [50] P. M. Ravdin, L. A. Simino,ff G. J. Davis et al., “Computer pro- gram to assist in making decisions about adjuvant therapy for women with early breast cancer,” Journal of Clinical Oncology, vol. 19, no. 4, Article ID 11181660, pp. 980–991, 2001. [51] E. Sarnowska, E. A. Grzybowska, K. Sobczak et al., “Hairpin structure within the 3’UTR of DNA polymerase beta mRNA acts as a post-transcriptional regulatory element and interacts with Hax-1,” Nucleic Acids Research,vol.35, no.16, ArticleID 17704138, pp. 5499–5510, 2007. [52] F. Zheng, C. Yue, G. Li et al., “Nuclear AURKA acquires kinase- independent transactivating function to enhance breast cancer stem cell phenotype,” Nature Communications,vol.7,no.10180, Article ID 26782714, 2016. MEDIATORS of INFLAMMATION The Scientific Gastroenterology Journal of World Journal Research and Practice Diabetes Research Disease Markers Hindawi Hindawi Publishing Corporation Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 http://www www.hindawi.com .hindawi.com V Volume 2018 olume 2013 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 International Journal of Journal of Immunology Research Endocrinology Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Submit your manuscripts at www.hindawi.com BioMed PPAR Research Research International Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 Journal of Obesity Evidence-Based Journal of Journal of Stem Cells Complementary and Ophthalmology International Alternative Medicine Oncology Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2013 Parkinson’s Disease Computational and Behavioural Mathematical Methods AIDS Oxidative Medicine and in Medicine Neurology Research and Treatment Cellular Longevity Hindawi Hindawi Hindawi Hindawi Hindawi www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018 www.hindawi.com Volume 2018

Journal

Journal of Oncology – Hindawi Publishing Corporation

Published: Apr 10, 2019

There are no references for this article.