Role of miR-139 as a surrogate marker for tumor aggression in breast cancer Hongyan Dai MD, PhD, Dan Gallagher BS, Sarah Schmitt BS, Ziyan Y. Pessetto PhD, Fang Fan MD, PhD, Andrew K. Godwin PhD, Ossama Tawfik MD, PhD PII: DOI: Reference:
S0046-8177(16)30311-2 doi: 10.1016/j.humpath.2016.11.001 YHUPA 4068
To appear in:
Human Pathology
Received date: Revised date: Accepted date:
22 June 2016 1 November 2016 3 November 2016
Please cite this article as: Dai Hongyan, Gallagher Dan, Schmitt Sarah, Pessetto Ziyan Y., Fan Fang, Godwin Andrew K., Tawfik Ossama, Role of miR-139 as a surrogate marker for tumor aggression in breast cancer, Human Pathology (2016), doi: 10.1016/j.humpath.2016.11.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
T
Title: Role of miR-139 as a surrogate marker for tumor aggression in breast cancer
RI P
Running head: miR-139: a marker for tumor aggression in breast cancer
SC
Hongyan Dai, M.D., Ph.D., Dan Gallagher, B.S., Sarah Schmitt, B.S., Ziyan Y. Pessetto, Ph.D., Fang Fan, M.D., Ph.D., Andrew K. Godwin, Ph.D., Ossama Tawfik, M.D., Ph.D.
Address correspondences to:
PT
Ossama Tawfik, MD, PhD
ED
MA
NU
Departments of Pathology and Laboratory Medicine, the University of Kansas Medical Center, Kansas City, Kansas 66160
Department of Pathology and Laboratory Medicine
CE
The University of Kansas Medical Center
AC
3901 Rainbow Boulevard Kansas City, KS 66160 Tel: (913) 588-1187
Fax: (913) 588-8780 Email:
[email protected] Conflict of Interest: “None of the authors have a conflict of interest either financial or of personal nature that is connected to the submitted manuscript". This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Ziyan Pessetto is funded by a CTSA grant from NCATS awarded to the University of Kansas Medical Center for Frontiers: The Heartland Institute for Clinical and Translational Research # KL2TR000119. The contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or NCATS. 1
ACCEPTED MANUSCRIPT Abstract MicroRNAs (miRNAs) are non-protein coding molecules that play a key role in oncogenesis,
T
tumor progression and metastasis in many types of malignancies including breast cancer. In the
RI P
current study, we studied the expression of miRNA-139-5p (miR-139) in invasive ductal carcinoma (IDC) of the breast and correlated its expression with tumor grade, molecular subtype,
SC
hormonal status, human epidermal growth factor receptor 2 (HER2) status, proliferation index,
NU
tumor size, lymph node status, patient's age and overall survival in 74 IDC cases. In addition, we compared and correlated miR-139 expression in 18 paired serum and tissue samples from
MA
patients with IDC to assess its value as a serum marker. Our data showed that miR-139 was down-regulated in all tumor tissue samples compared to control. More pronounced down
ED
regulation was seen in tumors that were higher grade, estrogen receptor negative, progesterone
PT
receptor negative, more proliferative or larger in size (P<0.05). Though not statistically significant, lower miR-139 level was frequently associated with HER2 overexpression.
CE
Additionally, significantly lower miR-139 tissue level was seen in patients who were deceased
AC
(P=0.027), though older age (>50 years) and positive local nodal disease did not adversely affect miR-139 expression. In contrast, serum miR-139 profile of the patients appeared similar to that of normal control. In conclusion, our study demonstrated that down regulation of miR-139 was associated with aggressive tumor behavior and disease progression in breast cancer. miR-139 may serve as a risk assessment biomarker in tailoring treatment options. Keywords: Breast cancer; invasive ductal carcinoma; micro RNA; microRNA-139
2
ACCEPTED MANUSCRIPT 1. Introduction The heterogeneity of breast cancer is well established with several molecularly distinct tumor
T
phenotypes characterized by gene expression microarray analysis. These include luminal A,
RI P
luminal B, normal breast-like, human epidermal growth factor receptor 2 (HER2) positive, and basal-like subtypes. Many tumors are known to respond to hormonal therapy and standard
SC
chemotherapy with excellent outcomes. However, a significant number of patients develop either
NU
regional or distant metastasis and eventually succumb to the disease.
MA
Many studies have demonstrated the significance of evaluating genetic signature of tumors, in addition to histopathologic parameters, in predicting recurrence, metastasis and survival [1]. The
ED
discovery of microRNAs (miRNAs) has created new avenues to the better understanding of the complexity and pathogenesis of breast cancer [2-8]. miRNAs are short non-coding RNAs that
PT
regulate gene expression at the translational level by inhibiting messenger RNA (mRNA)
CE
translation and, less frequently, by inducing mRNA degradation. They play a key role in oncogenesis, tumor progression and metastasis and exhibit the potential to act as either
AC
oncogenes or tumor suppressors in a given cellular context [9]. Dvinge et al. postulated that miRNAs exert their effect by acting as co-repressors or co-activators to modulate relationships between effector and target mRNAs, leading to differential co-expression of mRNAs [2]. While the overexpression of certain miRNAs has been shown to promote metastasis, the upregulation of others correlated with suppressed metastatic activity [3]. McDermott et al., have recently identified deregulation of 4 miRNAs (miR-29a, miR-181a, miR-223 and miR-652) in the circulation of women with Luminal A breast cancer [10]. Another study has proposed the use of miRNA as surrogate markers for circulating tumor cells and therefore as prognostic markers for metastatic breast cancer [4]. The expression of these miRNAs has the potential in not only 3
ACCEPTED MANUSCRIPT facilitating accurate subtyping of breast cancer but also in decision making for appropriate therapy.
RI P
T
miRNA-139-5p (miR-139), located at 11q13.4, has been shown to exhibit anti-oncogenic and anti-metastatic activities. Its dysregulation is associated with several types of carcinomas [11-
SC
13]. Several studies have demonstrated a vital role of miR-139 in breast cancer [5, 6]. Krishnan et al. postulated that miR-139 is crucial in regulating tumor cell invasion and migration in breast
NU
cancer [6]. Hua et al. demonstrated its ability in suppressing proliferation of Luminal type breast
MA
cancer cells by targeting Topoisomerase II alpha [5]. We hypothesized that miR-139 played a key role in tumor progression and metastasis in breast cancer. Our goal was to: 1) determine the
ED
expression level of miR-139 in formalin-fixed paraffin-embedded (FFPE) breast tumor tissue and correlate its level with known clinical and histopathological parameters; and 2) compare and
PT
correlate miR-139 expression in paired serum and tissue samples from breast cancer patients to
AC
CE
investigate the potential role of miR-139 as a circulating surrogate marker for tumor aggression.
4
ACCEPTED MANUSCRIPT 2. Materials and Methods
T
2.1 Patient Cohort:
RI P
This retrospective study was approved by the institutional review committee at the University of Kansas Medical Center. In order to determine miR-139 level in the FFPE breast cancer tissue, a
SC
total of 74 primary breast carcinomas diagnosed between 1997 (when HER2 testing became
NU
available) and 2010, for which estrogen receptor (ER), progesterone receptor (PR) and HER2 status and follow-up information was available, were examined. The samples were taken from 31
MA
mastectomy specimens, 32 lumpectomies and 3 incisional biopsies. Tumor miR-139 level was normalized to internal control SNORD 95. Histopathologic parameters, including histologic
ED
grade and type, hormonal and HER2 status, and proliferation index were extracted from patient pathology records. All tumors were graded using the modified "Nottingham" histological scoring
PT
system. Examples of tumor grading are depicted in Figure 1. Additional parameters including
CE
patients' age, tumor size and lymph node status were also recorded. Patient outcome data
AC
including survival status was obtained from the electronic medical records. Overall (rather than disease-specific) survival status was reported because discrepancies and/or incomplete data on cause of death were common. In order to determine the significance of serum miR-139 expression in breast cancer, 18 patients diagnosed with IDC from 2008 and 2012 were selected. miR-139 expression level in each sample (serum and FFPE) was compared to the mean of miR-139 expression level from 10 healthy donor serum and 10 unremarkable FFPE breast tissue samples, respectively, to yield a fold change. Tumor grade and size, ER/PR/HER2 status, proliferation index, and patient outcome were recorded. 5
ACCEPTED MANUSCRIPT 2.2 Immunohistochemistry: At diagnosis, tissue blocks containing the most representative and well-preserved tumor were
RI P
T
selected for immunohistochemistry (IHC). Immunohistochemical analysis was performed on tissue fixed with 10% neutral buffered formalin. IHC for tumor proliferation index (determined
SC
by Ki-67 immunostaining), ER, PR, and HER2 were performed at the time of diagnosis on all specimens. To determine Ki-67 labeling, the percentage of nuclei with immunopositivity was
NU
determined using the Proliferation Index program at first, the CAS (Cell Analysis System) 200
MA
image analyzer (Bacus Laboratory, Chicago, IL) prior to 2001 and the Automated Cellular Imaging System (ACISTM) (San Juan Capistrano, CA) thereafter. For ER and PR both the CAS-
ED
200 and the ACISTM systems were used for scoring. For HER2, ACISTM system was used. Positivity for ER and/or PR was defined as greater than 1% nuclear staining. Positive (3+) HER2
PT
staining was defined as greater than 10% strong membranous staining, per scoring instructions
CE
included in the HercepTest kit (DAKO, Carpinteria, CA). Fluorescence in-situ hybridization
rendered.
AC
(FISH) testing for HER2 amplification was performed whenever equivocal results (2+) were
2.3 miRNA sample preparation For the first cohort, total RNA was isolated from archival FFPE samples using miRNeasy FFPE kit (Qiagen) per manufacturer’s instructions. The concentration of RNA samples was detected by NanoQuant Infinite M200 Pro (Tecan). For the second cohort, paired archival FFPE and serum samples were obtained from 18 patients diagnosed IDC obtained from the Cancer Center Biospecimen Shared Resource Facility Repository Core Facility at the University of Kansas Medical Center. For total RNA isolation 6
ACCEPTED MANUSCRIPT from FFPE samples, the miRNeasy FFPE kit (Qiagen) was again used according to manufacturer instructions, and concentrations determined using the NanoQuant Infinit M200 Pro (Tecan). A C.
T
elegans miRNA mimic cel-miR-39 (Qiagen) was added to each FFPE-derived RNA sample at a
RI P
concentration of 108 copies per 20 ng of total RNA to facilitate assessment of the presence of inhibitors. For total RNA isolation from serum, frozen archival serum samples were thawed on
SC
ice, then total RNA isolated from 200 μl serum using a miRNeasy Serum/Plasma Kit (Qiagen)
NU
following manufacturer instructions. Cel-miR-39 (Qiagen) was added to each sample during isolation, immediately following the addition of denaturing solution, according to manufacturer
MA
instructions for internal normalization. RNA concentrations were determined using the NanoQuant Infinite M200 Pro (Tecan). All samples were immediately stored and kept at -80 °C
ED
until use.
PT
2.4 RT-PCR
CE
The miScript II RT kit (Qiagen) was used to synthesize first-strand cDNA from an equal amount
AC
of (100 ng) of RNA for all FFPE samples, and from equal volumes (2 μl) of RNA for all serum samples. The reverse-transcription reaction was conducted using the Hi-Spec buffer provided in the miScript II RT Kit in a total volume of 20 μl, according to manufacturer instructions. Each cDNA sample was diluted with 200 μl RNase-free H2O and stored at -20 °C until further use. The miR-139 specific primers and the SNORD 95 primers were used for samples in the first cohort in this study. Expression levels were normalized against endogenous SNORD 95 control for the individual assay. Expression levels of mature miR139 and SNORD 95 were determined by real-time quantitative polymerase chain reaction (q-PCR) using the miScript SYBR green PCR kit (Qiagen) in LightCycler 480 system (Roche) according to manufacturer instructions. 7
ACCEPTED MANUSCRIPT For determination of miR-139 levels in paired archival FFPE and serum samples in the second cohort, the miScript SYBR green PCR kit with the LightCycler 480 system was used for
T
amplification of target miR-139, cel-miR-39 internal normalizer, RNU6b for FFPE samples, and
RI P
SNORD 95 for serum samples. Data was normalized to cel-miR-39 spike-in control calibration then either RNU6b for FFPE or SNORD 95 for serum. The relative expression of miR-139 in
SC
tumor tissue compared to mean normal tissue expression for each sample type (serum or FFPE)
NU
was calculated using the comparative cycle threshold (ΔΔCt) method, yielding a fold-change in expression relative to normal samples. Fold up- or down- regulation was then calculated from
MA
fold-change values to represent fold-change results in a biologically meaningful way, with foldchange values greater than one indicating a positive- or an up-regulation and fold-change values
PT
2.5 Statistical Analysis:
ED
less than one indicating a negative or down-regulation.
CE
Overall frequencies and percentages were summarized for tumor grade, histology, lymph node status, and evidence of expression by IHC for ER, PR, and HER2 (and thus TN status), as well as
AC
Ki-67. Data was reported as mean ± SD. Values were compared using the Student T-test or with one-way ANOVA when three or more groups were present using SigmaPlot software (Systat) or with GraphPad version 5.04 for Windows (GraphPad Software). Survival analysis was performed using GraphPad Software. A P value less than 0.05 was considered to be statistically significant.
8
ACCEPTED MANUSCRIPT 3. Results
T
Differential expression of miR-139 in breast cancer and its correlation with tumor
RI P
aggression:
Table 1 summarizes the clinicopathologic parameters for the 74 patients included in the first
SC
cohort. The mean age of patients was 57 (+14.2) years. Sixty-six percent of the patients were 50
NU
years or older. When the miR-139 expression levels were normalized against endogenous SNORD 95 control for the individual assay, miR-139 was differentially expressed in breast
MA
cancers (Figures 2 & 3). There was a statistically significant correlation between under expression of miR-139 and several well established adverse histopathologic and clinical
ED
parameters including high tumor grade (Figure 2A, P<0.05), basal-like/triple negative molecular subtype (Figure 2B, P<0.05), loss of hormonal receptor expression (Figure 3A & B, P<0.01),
PT
higher proliferative activity (Ki-67>15%, Figure 3D, P=0.0002). Although not statistically
CE
significant, more pronounced reduction of miR-139 was observed in larger tumors (>2 cm, Figure 3E, P=0.0708) and tumors with HER2 overexpression (Figure 3C, P=0.0537). When
AC
patient’s age at initial diagnosis was examined, no difference in miR-139 level was identified in patients older than 50 years old versus those who were younger (Figure 3F, P=0.434). When levels of miR-139 expression were correlated with overall survival, significantly lower miR-139 was observed in deceased patients compared to those who were alive with a follow-up time ranging from 5 to 18 years (Figure 3I, P=0.027). Although no significant difference in miR-139 expression was identified in tumors with positive regional lymph node metastasis versus those without in the entire cohort of patients (Figure 3G, P=0.264), significantly more attenuated miR139 levels were seen in tumors with nodal disease (Figure 3H, P =0.0264) when only luminal A 9
ACCEPTED MANUSCRIPT cases were included for the comparison. When patient survival was examined in the first cohort, patients with fold down regulation greater than or equal to the median fold change in tissue miR-
T
139 expression did not differ significantly from those with less decrease (Figure 4, P>0.05,
RI P
median fold change = -1.006).
SC
Expression of miR-139 in breast cancer tumor tissue and corresponding serum In the second cohort, matching serum and FFPE tumor tissue samples from a total of 18 patients
NU
were examined. The demographic and histopathologic parameters of these patients are
MA
summarized in Table 2. Significant down regulation (8.7-fold, P<0.05) of miR-139 was observed in all FFPE tumor tissue samples relative to normal breast tissue controls (Figure 5). However,
ED
corresponding serum samples revealed down regulation of miR-139 in only 14 of 18 samples (Figure 6). We further confirmed the findings above and showed that loss of miR-139 expression
PT
was more pronounced in IDCs with aggressive pathologic and clinical features in tumor tissue.
CE
The statistically significant decrease of miR-139 expression was associated with tumors of larger size (>2cm, Figure 7A, P=0.014) and higher proliferative index (Ki67 >20%, Figure 7B,
AC
P=0.012). Though not statistically significant, lower miR-139 was also seen in high grade tumors (grade III vs. grade I and II, Figure 7C, P=0.214). When miR-139 serum level was correlated with the above parameters, no significant difference was observed. When patient survival was examined, those with equal or more than median fold decrease in tissue miR-139 level did not differ significantly from patients with less decrease (Figure 8a, P>0.05). When serum miR-139 level was correlated with patient survival, patients with miR-139 down regulation was comparable to those with up regulation (Figure 8b, P>0.05).
10
ACCEPTED MANUSCRIPT 4. Discussion:
T
Breast cancer is a global health threat for women and one of the leading causes of cancer
RI P
mortality [14]. A group of clinical and pathological biomarkers is being used clinically to classify breast cancer including tumor size, histological grade, hormone receptor status, HER2
SC
amplification, proliferation rate and lymph node status. Although all these parameters have
NU
important clinical values, they are not always predictive of tumor behavior and clinical outcome. Also, with the increasing awareness and better screening modalities, more breast cancer cases are
MA
diagnosed at earlier stages where challenge arise as to identifying patients who will likely benefit from adjuvant therapies versus those who can be adequately treated with surgery alone. Further
ED
insight into the molecular mechanisms underlying breast cancer is warranted for the
PT
identification of additional diagnostic and prognostic biomarkers.
CE
miRNAs are small non-coding highly conserved RNA molecules regulating gene expression at the translational level by silencing target mRNAs. miRNAs are a powerful modulator of mRNAs
AC
with a single miRNA capable of targeting hundreds of mRNAs. Since its discovery, aberrations of miRNA expression have been found to play a vital role in virtually all types of cancers. [12, 15-19] The importance of miRNA deregulation in breast cancer has been demonstrated and reiterated by many studies. Iorio and colleagues performed a genome-wide miRNA expression profiling in normal and breast tumor tissues and identified 15 miRNAs that are either consistently over or under expressed with likely downstream targets being oncogenes or tumor suppressor genes, respectively [7]. Ma et al. revealed that overexpression of miR-10b promotes robust invasion and metastasis in breast cancer and its level is positively correlated with tumor size, histologic grade, clinical staging, lymph node metastasis, tumor proliferation, and HER2 11
ACCEPTED MANUSCRIPT amplification [8]. Also, overexpression of miR-21 and underexpression of miR-205 have been
T
associated with shorter disease-free interval in early breast cancer patients [3].
RI P
miR-139 exhibits anti-oncogenic and anti-metastatic activities in many types of carcinomas [1113]. A previous study reported that miR-139 is significantly down-regulated in breast cancer
SC
tissue compared to adjacent normal controls [6]. Additionally, a human invasive breast cancer cell line, MDA-MB-231, manipulated to overexpress miR-139 demonstrated reduced invasive
NU
and migratory abilities [6]. In the current study, we demonstrated that miR-139 levels are lower
MA
in breast tumors, confirming the previous findings by Krishnan et al. [6]. We further concluded that the extent of down-regulation was significantly associated with aggressive tumor behavior
ED
and poor clinical outcome (Figures 2, 3 and 7). Lower levels of miR-139 were also associated with a higher grade, hormonal receptor negative, highly proliferative tumors, and in patients with
PT
fatal outcome (Figure 2, 3 and 7). Evaluation of the relationship between its levels of expression
CE
in tumors with and without regional metastasis in lymph node was clinically interesting. When the entire cohort of tumors was collectively evaluated there was no significant difference in miR-
AC
139 expression in tumors with or without nodal metastasis (Figure 3G). That was not the case when luminal A tumors were studied. Significantly more pronounced loss of miR-139 was identified in tumors with nodal disease (Figure 3H). This is potentially very valuable in not only facilitating more accurate subtyping of breast tumor by conventionally known methods but also in decision making for appropriate therapy. Survival analysis (Figures 4, 8a) failed to demonstrate correlation between miR-139 down regulation and patient survival. These findings may be due to relative short followup duration.
12
ACCEPTED MANUSCRIPT In the current study, only IDC of no special type (NST) was included. IDC is the most commonly diagnosed breast cancer subtype and accounts for the vast majority of breast cancer
T
cases. We observed a decrease of miR-139 expression level in all cases. This is consistent with
RI P
previously reported findings by Iorio et al. who revealed similar miRNA expression profile in ductal and lobular carcinomas [7]. A number of relatively rare subtypes of breast cancer have
SC
been established which are composed of an extremely heterogeneous group of tumors. The
NU
clinical behavior of these rare tumors is not always in concordance with their grading. For example, adenoid cystic carcinoma is a clinically indolent tumor with excellent prognosis despite
MA
falling into ER, PR, and HER2 negative and basal-like category. Medullary carcinoma, on the other hand, is believed to carry a better prognosis in its pure form despite being a high grade
ED
tumor. It has been shown; however, some subtypes with worst prognosis, including metaplastic
PT
carcinoma and atypical medullary carcinoma, exhibit more profound loss of miR-139 than IDC
CE
of NST [6].
A host of downstream targets of miR-139 have been identified including, FoxO1 [20], rho-
AC
kinase2 [19], and c-Fos [21] in hepatocellular carcinoma, CXCR4 [17] in gastric carcinoma, RAP1B [18] and type I insulin-like growth factor [22] in colorectal cancer, among others. In breast carcinoma, possible pathways that miR-139 affects include Wnt signaling pathways, receptor tyrosine kinases signaling through several pathways including RAS-MAPK and PI3 kinase, and TGFβ signaling pathways [6]. These downstream targets have been shown to play important roles in breast cancer progression, invasion, and metastasis [6, 23-25].
The importance of extracellular miRNAs present in the bloodstream and body fluids (circulating miRNA) in cancer was first reported by Lawrie and colleagues who demonstrated elevated 13
ACCEPTED MANUSCRIPT serum miRNA-21 level in patients with large B-cell lymphomas [16]. Recent advances in bloodbased miRNA profiling studies have generated a concept that circulating cancer specific
T
biomarkers holds great potential as a minimally invasive marker in the initial diagnosis, outcome
RI P
prognostication, therapy response monitoring, and recurrence surveillance of cancer. Since its first description, several studies have addressed the role of circulating miRNAs as a molecular
SC
biomarker in various types of carcinomas [15, 26, 27]. Zhu et al. discovered that in PR-positive
NU
breast tumors there was higher serum level of miRNA-155 than PR-negative ones. Also, Heneghan and colleagues demonstrated that the levels of miRNA-195 and let-7a decreased
MA
postoperatively to level comparable with control [28]. Furthermore, circulating miRNA-21 has been shown to be differentially expressed in breast cancer patients versus healthy control and can
ED
distinguish patients with distant metastasis from those with only local disease [29]. Despite
PT
recent advances in molecular biology and cancer research, a clinically validated systemic breast cancer biomarker, regardless of cancer subtypes, grades, and hormone status, has not been
CE
identified. In the second cohort, we evaluated the role of miR-139, by itself, as a potential serum
AC
disease marker for breast cancer. Unfortunately, our cohort failed to demonstrate eithera unanimous down-regulation of miR-139, as observed in FFPE tissue, or apparent correlation between tissue and serum miR-139 level. The above findings indicate that serum miR-139 alone may not be a reliable indicator for breast cancer monitoring/screening. More studies need to be undertaken in search for other potential clinically useful markers. In summary, our study showed that miR-139 is differentially expressed in invasive breast carcinomas; its down-regulation is significantly associated with aggressive tumor behavior and disease progression. miR-139 may potentially serve as a new risk assessment biomarker in tailoring treatment options in breast cancer patients. This study confirms that value of miRNAs 14
ACCEPTED MANUSCRIPT as potential prognostic and predictive biomarkers and identify miR-139 for additional development as a clinically relevant biomarker to guide treatment of aggressive and advanced
AC
CE
PT
ED
MA
NU
SC
RI P
T
breast carcinomas.
15
ACCEPTED MANUSCRIPT Acknowledgements The authors would like to acknowledge the support of the Biospecimen Repository Core Facility
T
staff funded in part by the University of Kansas Cancer Center (P30 CA168524). AKG is the
AC
CE
PT
ED
MA
NU
SC
RI P
Chancellors Distinguished Chair in Biomedical Sciences Endowed Professor.
16
ACCEPTED MANUSCRIPT References: Aftimos PG, Barthelemy P, Awada A. Molecular biology in medical oncology: diagnosis, prognosis, and precision medicine. Discov Med 2014;17:81-91.
[2]
Dvinge H, Git A, Gräf S, et al. The shaping and functional consequences of the microRNA landscape in breast cancer. Nature 2013;497:378-82.
[3]
Markou A, Yousef GM, Stathopoulos E, Georgoulias V, Lianidou E. Prognostic significance of metastasis-related microRNAs in early breast cancer patients with a long follow-up. Clin Chem 2014;60:197-205.
[4]
Madhavan D, Peng C, Wallwiener M, et al. Circulating miRNAs with prognostic value in metastatic breast cancer and for early detection of metastasis. Carcinogenesis 2016;37:461-70.
[5]
Hua W, Sa KD, Zhang X, et al. MicroRNA-139 suppresses proliferation in luminal type breast cancer cells by targeting Topoisomerase II alpha. Biochem Biophys Res Commun 2015;463:1077-83.
[6]
Krishnan K, Steptoe AL, Martin HC, et al. miR-139-5p is a regulator of metastatic pathways in breast cancer. RNA 2013;19:1767-80.
[7]
Iorio MV, Ferracin M, Liu CG, et al. MicroRNA gene expression deregulation in human breast cancer. Cancer Res 2005;65:7065-70.
[8]
Ma L. Role of miR-10b in breast cancer metastasis. Breast Cancer Res 2010;12:210.
[9]
Kent OA, Mendell JT. A small piece in the cancer puzzle: microRNAs as tumor suppressors and oncogenes. Oncogene 2006;25:6188-96.
[10]
McDermott AM, Miller N, Wall D, et al. Identification and validation of oncologic miRNA biomarkers for luminal A-like breast cancer. PLoS One 2014;9: e87032.
[11]
Xu W, Hang M, Yuan CY, Wu FL, Chen SB, Xue K. MicroRNA-139-5p inhibits cell proliferation and invasion by targeting insulin-like growth factor 1 receptor in human non-small cell lung cancer. Int J Clin Exp Pathol 2015;8:3864-70.
[12]
Yue S, Wang L, Zhang H, et al. miR-139-5p suppresses cancer cell migration and invasion through targeting ZEB1 and ZEB2 in GBM. Tumour Biol 2015;36:6741-9.
[13]
Zhang HD, Sun DW, Mao L, et al. MiR-139-5p inhibits the biological function of breast cancer cells by targeting Notch1 and mediates chemosensitivity to docetaxel. Biochem Biophys Res Commun 2015;465:702-13.
AC
CE
PT
ED
MA
NU
SC
RI P
T
[1]
17
ACCEPTED MANUSCRIPT Jemal A, Center MM, DeSantis C, Ward EM. Global patterns of cancer incidence and mortality rates and trends. Cancer Epidemiol Biomarkers Prev 2010;19:1893-907.
[15]
Feng G, Li G, Gentil-Perret A, Tostain J, Genin C. Elevated serum-circulating RNA in patients with conventional renal cell cancer. Anticancer Res 2008;28:321-6.
[16]
Lawrie CH, Gal S, Dunlop HM, et al. Detection of elevated levels of tumourassociated microRNAs in serum of patients with diffuse large B-cell lymphoma. Br J Haematol 2008;141:672-5.
[17]
Bao W, Fu HJ, Xie QS, et al. HER2 interacts with CD44 to up-regulate CXCR4 via epigenetic silencing of microRNA-139 in gastric cancer cells. Gastroenterology 2011;141:2076-2087.e6.
[18]
Guo H, Hu X, Ge S, Qian G, Zhang J. Regulation of RAP1B by miR-139 suppresses human colorectal carcinoma cell proliferation. Int J Biochem Cell Biol 2012;44:146572.
[19]
Wong CC, Wong CM, Tung EK, et al. The microRNA miR-139 suppresses metastasis and progression of hepatocellular carcinoma by down-regulating Rho-kinase 2. Gastroenterology 2011;140:322-31.
[20]
Hasseine LK, Hinault C, Lebrun P, Gautier N, Paul-Bellon R, Van Obberghen E. miR139 impacts FoxO1 action by decreasing FoxO1 protein in mouse hepatocytes. Biochem Biophys Res Commun 2009;390:1278-82.
[21.
Fan Q, He M, Deng X, et al. Derepression of c-Fos caused by microRNA-139 downregulation contributes to the metastasis of human hepatocellular carcinoma. Cell Biochem Funct 2013;31:319-24.
[22]
Shen K, Liang Q, Xu K, et al. MiR-139 inhibits invasion and metastasis of colorectal cancer by targeting the type I insulin-like growth factor receptor. Biochem Pharmacol 2012;84:320-30.
[23]
Scheel C, Eaton EN, Li SH, et al. Paracrine and autocrine signals induce and maintain mesenchymal and stem cell states in the breast. Cell 2011;145:926-40.
[24]
Padua D, Zhang XH, Wang Q, et al. TGFbeta primes breast tumors for lung metastasis seeding through angiopoietin-like 4. Cell 2008;133:66-77.
[25]
Kim J, Shao Y, Kim SY, et al. Hypoxia-induced IL-18 increases hypoxia-inducible factor-1alpha expression through a Rac1-dependent NF-kappaB pathway. Mol Biol Cell 2008;19:433-44.
AC
CE
PT
ED
MA
NU
SC
RI P
T
[14]
18
ACCEPTED MANUSCRIPT Ng EK, Chong WW, Jin H, et al. Differential expression of microRNAs in plasma of patients with colorectal cancer: a potential marker for colorectal cancer screening. Gut 2009;58:1375-81.
[27]
Chen X, Ba Y, Ma L, et al. Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and other diseases. Cell Res 2008;18:997-1006.
[28]
Heneghan HM, Miller N, Lowery AJ, Sweeney KJ, Newell J, Kerin MJ. Circulating microRNAs as novel minimally invasive biomarkers for breast cancer. Ann Surg 2010;251:499-505.
[29]
Asaga S, Kuo C, Nguyen T, Terpenning M, Giuliano AE, Hoon DS. Direct serum assay for microRNA-21 concentrations in early and advanced breast cancer. Clin Chem 2011;57:84-91.
AC
CE
PT
ED
MA
NU
SC
RI P
T
[26]
19
ACCEPTED MANUSCRIPT Figure Legends Figure 1. Examples of invasive ductal carcinoma of the breast and their grading. A,
T
Infiltrative breast cancer demonstrates prominent ductular formation, low mitotic activity and
RI P
monomorphous nuclei, consistent with modified Nottingham histological grade I. B, Invasive breast cancer cells are relatively monophasic but arranged in solid nests. This tumor is
SC
considered to be modified Nottingham histological grade II. C, Tumor demonstrates solid
NU
infiltrate of neoplastic cells with marked nuclear pleomorphism and brisk mitotic activity (inset) and is consistent with modified Nottingham grade III. (hematoxylin-eosin stain, original
MA
magnifications x 200)
ED
Figure 2. The extent of down regulation of miR-139 in tumor tissue correlated with tumor
PT
grade and molecular tumor subtypes. Significant attenuation of miR-139 expression was seen in grade II tumors (P>0.001 vs. grade I, A) and grade III tumors (P=0.022 vs. grade II; P<0.001
CE
vs. grade I, A). Significant difference in miR-139 expression was also seen in different molecular
AC
subtypes of breast cancer (P<0.05, B).
Figure 3. Down regulation of miR-139 correlated with aggressive histopathologic parameters and adverse clinical outcome . More significant under expression of miR-139 was seen in ER negative (P=0.0008, A), PR negative (P=0.005, B), high proliferative (Ki-67>15%, P=0.0002, D) tumors and in tumors from deceased patients (P=0.027, I). Although there was no correlation between miR-139 expression and local lymph node metastasis overall (P=0.264, G), significant correlation was noted in Luminal A tumors with known regional metastasis compare to non-metastatic tumors. (P=0.0264, H). 20
ACCEPTED MANUSCRIPT
Figure 4: Patient survival in correlation with miR-139 tissue level (1st cohort). Patients with
T
a fold decrease greater than or equal to the median fold change in tissue miR-139 expression
RI P
observed in the first cohort did not show a significant difference from survival of patients with
SC
an observed decrease less than the median value (P>0.05, median fold change = -1.006).
NU
Figure 5: miR-139 was down regulated in IDC tumor tissue. miR-139 expression levels in each FFPE tumor tissue was compared to a set of 10 breast tissue samples with no diagnostic
MA
abnormalities (Normal controls). Results showed significant decrease in miR-139 level in tumor
ED
tissue (P=0.037).
PT
Figure 6. miR-139 expression levels in 18 paired serum and tumor tissue samples. miR-139 expression levels were measured by in FFPE tumor tissue and matching serum samples from
CE
breast cancer patients (n=18). These miR-139 levels were then normalized to the levels
AC
established in benign breast tissue (n=10) or serum samples from healthy control patients (n=10), respectively. Fold-change values greater than one indicate a positive- or an up-regulation. Corresponding ER, PR, and HER2 status is listed in the table below. Fold-change values less than zero indicate down-regulation. Down-regulation of miR-139 was observed in all IDC tumors tissues; however, corresponding serum samples showed varied miR-139 expression.
Figure 7. Lower levels of miR-139 were associated with aggressive pathologic and clinical features in tumor tissue (2nd cohort). A statistically significant decrease (P<0.05) of miR-139 levels was observed in tumors with larger size (>2cm, A) and higher proliferative index (Ki67 21
ACCEPTED MANUSCRIPT >20%, B) compared to benign breast tissue. Though not statistically significant, lower miR-139
RI P
T
levels were seen in grade III tumors compare to grade I and II (P=0.214, C).
Figure 8. Patient survival in correlation with miR-139 level (2nd cohort). Patients with equal
SC
or more than the median decrease in tissue miR-139 level observed in the second cohort did not differ significantly from patients with less decrease (A, P>0.05, median fold change = -9.932).
NU
When serum miR-139 level was correlated with patient survival, patients with miR-139 down
AC
CE
PT
ED
MA
regulation was comparable to those with up regulation (B, P>0.05).
22
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
Figure 1
23
AC
Figure 2a
CE
PT
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
24
AC
Figure 2b
CE
PT
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
25
Figure 3
AC
CE
PT
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
26
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 4
27
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
Figure 5
28
AC
Figure 6
CE
PT
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
29
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 7a
30
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 7b
31
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 7c
32
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 8a
33
ED
MA
NU
SC
RI P
T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 8b
34
ACCEPTED MANUSCRIPT Table 1: Distribution of clinical and histopathologic parameters in breast cancer patients. Clinical feature or frequency of biomarker expression
CE
PT
RI P NU
SC
ED
MA
Grade (SBR) I II III Molecular subtype Luminal A Luminal B Basal-like HER2 Lymph Node Status Positive Negative Survival Status Alive Deceased Not known ER Positivity, ≥1% No Yes PR Positivity, ≥1% No Yes HER2 Positivity No Yes High Ki-67 (expression, ≥ 20% ) No Yes
AC
25 (34%) 49 (66%)
T
Age at diagnosis <50 years >50 years Tumor size, cm < 2 cm >2 cm Surgical Procedure Incisional Biopsy Lumpectomy Mastectomy Histology (Invasive Ductal)
Total Number
33 (45%) 41 (55%) 3 (4%) 40 (54%) 31 (42%) 74 11 (15%) 27 (36%) 36 (49%) 29 (39%) 21 (29%) 20 (27%) 4 (5%) 27 33 55 (74%) 12 (15%) 7 (11%) 50 (68%) 24 (32%) 48 (65%) 26 (35%) 64 (87%) 10 (13%) 40 (54%) 34 (46%) 35
ACCEPTED MANUSCRIPT Table 2. Distribution of clinical and histopathologic parameters in breast cancer patients.
Total Number
T
Clinical feature or frequency of biomarker expression Age at diagnosis <50 years >50 years Tumor size, cm < 2 cm >2 cm Surgical Procedure Lumpectomy Mastectomy Histology (Invasive Ductal)
AC
CE
PT
ED
MA
Grade (SBR) I II III Molecular subtype Luminal A Luminal B Basal-like HER2 Lymph Node Status Positive Negative Survival Status Alive Deceased ER Positivity, ≥1% No Yes PR Positivity, ≥1% No Yes HER2 Positivity No Yes High Ki-67 (expression, ≥ 20% ) No Yes
NU
SC
RI P
2 (11%) 16 (89%) 12 (67%) 6 (33%) 8 (44%) 10 (56%) 18
2 (11%) 10 (56%) 6 (33%) 9 (50%) 3 (28%) 4 (17%) 2 (6%) 11 7 9 (50%) 9 (50%) 5 (28%) 13 (72%) 6 (33%) 12 (67%) 16 (89%) 2 (11%) 11 (61%) 7 (39%)
37