Development and evaluation of a novel RT-qPCR based test for the quantification of HER2 gene expression in breast cancer
Hicham El Hadi, Imane Abdellaoui-Maane, Denise Kottwitz, Manal El Amrani, Nadia Bouchoutrouch, Zineb Qmichou, Mehdi Karkouri, Hicham ElAttar, Hassan Errihani, Pedro L Fernandez, Youssef Bakri, Hassan Sefrioui, Abdeladim Moumen
PII: S0378-1119(16)31027-7
DOI: doi:10.1016/j.gene.2016.12.027
Reference: GENE 41725
To appear in: Gene
Received date: 16 June 2016 Revised date: 8 December 2016 Accepted date: 23 December 2016
Please cite this article as: Hicham El Hadi, Imane Abdellaoui-Maane, Denise Kottwitz, Manal El Amrani, Nadia Bouchoutrouch, Zineb Qmichou, Mehdi Karkouri, Hicham ElAttar, Hassan Errihani, Pedro L Fernandez, Youssef Bakri, Hassan Sefrioui, Abdeladim Moumen , Development and evaluation of a novel RT-qPCR based test for the quantification of HER2 gene expression in breast cancer. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Gene(2016), doi:10.1016/j.gene.2016.12.027
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
Development and evaluation of a novel RT-qPCR based test for the quantification of HER2 gene expression in breast cancer.
Hicham El Hadia,b, Imane Abdellaoui-Maanea,b, Denise Kottwitza, Manal El Amrania, Nadia Bouchoutroucha, Zineb Qmichoua, Mehdi Karkouric, Hicham ElAttard, Hassan Errihanie, Pedro.L Fernandezf, Youssef Bakrib, Hassan Sefriouia and Abdeladim Moumena*
aMedical biotechnology center,Moroccan Foundation for Advanced Science Innovation and Research (MASCIR), Rue Mohamed Al JazouliMadinat Al Irfane, Rabat, Morocco.
bBiochemistry and Immunology Laboratory, Faculty of Science Rabat University MohamedV, Avenue Ibn Battouta1014, Rabat, Morocco.
cPathology department, HOPITAL IBN ROCHD, quartier des hopitaux, Casablanca, Morocco.
dLaboratory of Pathology MlyIdriss I, Boulevard MoulayIdriss 1er Quartier Des Hopitaux, Casablanca, Morocco.
eNational Institute of Oncology (INO), Avenue Allal El Fassi, Rabat, Morocco.
fDepartment of Anatomical Pathology, Hospital Clínic and Institut d'Investigacions Biomèdiques August Pi iSunyer (IDIBAPS), University of Barcelona, Spain.
h.elhadi@mascir.com i.abdellaoui@mascir.com z.qmichou@mascir.com denisekottwitz@web.de dahmani_manale@yahoo.fr ybakri@gmail.com laboratoiremoulaydriss1er@gmail.com h.sefrioui@mascir.com plfernan@clinic.ub.es
mehdi.karkouri@menara.ma
*Corresponding authors
Abdeladim Moumen, MS, PhD. HDR Email :a.moumen@mascir.com Tel: 212-530 410 501
Fax: 212-530 279 827
ACCEPTED MANUSCRIPT
Abstract:
Accurate measurement of Human epidermal growth factor receptor (HER2) gene expression is central for breast or stomach cancer therapy orientation and prognosis. The current standards testing methods for HER2 expression are immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH).
In the current study, we explored the use of quantitative real time reverse transcription-PCR (RT-qPCR) as a potential method for the accurate relative quantification of the HER2 gene using formalin fixed paraffin embedded (FFPE) breast cancer biopsy samples. The main aim of the current study is to measure the level of concordance of RT-qPCR based quantification of HER2 overexperession with both IHC and FISH. Accordingly, an endogenous control gene (ECG) is required for this relative quantification and should ideally be expressed equivalently across tested samples. Stably expressed ECGs have been selected from a panel of seven genes using GenEx V6 software which is based on geNorm and NormFinder and statistical methods.
Quantification of HER2 gene expression was performed by our RT-qPCR-based test and compared to the results obtained by both IHC and FISH methods.
HER2 gene quantification using RT-qPCR test was normalized using the two ECGs (RPL30 and RPL37A) that were successfully identified and selected from a panel of seven genes as the most stable and reliable ECGs. We evaluated a total of 216 FFPE tissue samples from breast cancer patients. The results obtained with RT-qPCR in the current study were compared to both IHC and FISH data collected for the same patients. In addition to an internal evaluation, an external evaluation of this assay was also performed in a recognized pathology center in Europe (Clinic Barcelona Hospital Universitari, Spain) using 116 FFPE breast cancer tissue samples. The results demonstrated a high concordance between RT-qPCR and either IHC (98%) or FISH (72%) methods. Accordantly, the overall concordance was 85%.
To our knowledge, this is the first study using the specific combination of RPL30 and RPL37 as reference genes for an accurate HER2 gene quantification in FFPE biopsy samples.
Although further clinical validation regarding evolution and therapeutic response using RT- qPCR for the quantification of HER2 expression are still needed, the present study constitutes definitely a factual element that the RT-qPCR based assay may constitute a valid complementary test to accurately measure HER2 expression for a better treatment orientation.
KEY WORDS
Breast cancer, HER2 expression, RT-qPCR, Endogenous control genes, Molecular diagnosis.
ACCEPTED MANUSCRIPT
1. Introduction
The gene for the human epidermal growth factor receptor 2 (EGFR2/erbB2/HER2/neu) is a proto-oncogene located on chromosome 17 (17q21.1) and encodes for 185-kda transmembrane oncoprotein with tyrosin kinase activity (Andersen et al., 2004; Baselga et al., 2004). It is well established that HER2 is overexpressed in 15% to 30% of human breast cancers cases (Bast et al., 2001; Baselga et al., 2004; BOSSARD et al., 2005) which are often associated with a poor prognosis and a more aggressive cancer phenotype. Therefore, HER2 is considered as a prognostic marker for aggressive cancers and an important predictor of therapeutic response (Bast et al., 2001; BOSSARD et al., 2005). Indeed, in HER2 positive patients, combined treatment of chemotherapy with anti-HER2 antibodies (e.g. Trastuzumab), improved the treatment efficacy, the time of progression, the response rate and the survival rate when compared to chemotherapy alone (Slamon et al., 2001).
An accurate assessment of HER2-status for breast cancer patients is crucial since only patients with HER2 protein overexpression or gene amplification are eligible for trastuzumab treatment (Chang et al., 1998). However, a false-positive test outcome could lead to expensive and ineffective treatments associated with unnecessary side-effects and would affect improvement of both the response and survival rates compared to chemotherapy alone (Callahan, 1989; Bustin et al., 2005) and a false-negative test result will deprive the patient of an important and ideal target therapeutic option. HER2 gene amplification is tightly associated with mRNA overexpression and increased protein levels. Thus, HER2 alterations can be measured at the DNA, mRNA, and protein levels. Both immunohistochemical (IHC) and fluorescent in situ hybridization (FISH) are two approved diagnostic techniques by the US Food and Drug Administration for assigning HER2 status in the clinical laboratory based on outcome data and response to trastuzumab treatment (Drury et al., 2009). While IHC uses an antibody to evaluate HER2 protein expression, FISH determines the number of HER2 copies using a DNA probe coupled to a fluorescent, detection system (Hanna et al., 2014).
A combined IHC/FISH approach has become more popular than approaches using FISH alone. The FISH method is a high-tech, time-consuming, and expensive method. The results of this method can be ambiguous within the score range of 1.8- 2.2 (Gutierrez and Schiff, 2011). In addition, IHC testing is limited by technical considerations and the inherent subjectivity of scoring (Sauter et al., 2009). Although there is still debate about the best
ACCEPTED MANUSCRIPT
method to determine HER2 status (Wolff et al., 2007), currently there is no ‘gold standard’
method for HER2 testing (Kılıç et al., 2014).
Several studies have compared HER2 mRNA expression by reverse transcription quantitative polymerase chain reaction (RT-qPCR) to results obtained by both IHC and FISH (Livak and Schmittgen, 2001; Pazhoomand et al., 2013). Many of these studies have shown a high quantification concordance between the two methods suggesting that RT-qPCR could be used for the quantification of the HER2 overexpression in breast cancer patients as a simple, reproducible, cost effective and widely applicable method (Barberis et al., 2008). Therefore, RT-qPCR, constitutes an useful alternative diagnostic approach for HER2 status scoring in human breast cancer since it is insensitive to interobserver variability, automated and easily amenable to standardization represents, (Vinatzer et al., 2005; Barberis et al., 2008; Lehmann- Che et al., 2011). Nevertheless, there are still controversies over whether RT-qPCR could be well used for a better prognosis for antiHER2-based breast cancer therapy (Esteva et al., 2005; Perez et al., 2015; Yamamoto-Ibusuki et al 2015). Indeed, while some studies have shown that RT-qPCR could be used to accurately measure HER overexpression and breast cancer prognosis, others have shown that RT-qPCR has inadequate sensitivity, rendering it unsuitable to determine HER2 status (Esteva et al., 2005; Perez et al., 2015 ; Yamamoto- Ibusuki et al 2015). However, an ideal endogenous control gene (ECG) is mandatory for an accurate measurement of the relative quantification of HER2 expression,. The ECG is selected from the cellular genes that are expressed equivalently across all assessed samples.
Many so called housekeeping genes such as GAPDH, tubilin and beta-actin have been used as ECGs for the quantification of gene expression in cancerous cells. However, several studies have shown that most of these genes are not stably expressed between different cancer types (Rondinelli et al., 1997; Pegram et al., 1998; Press et al., 2002) and cannot be used for an accurate gene expression quantification. Therefore, statistical models and software programs have been developed for the analysis of candidate ECG gene stability.
In the present study, we first aimed to select the most stable ECGs suitable for an accurate evaluation of HER2 gene expression in human breast cancer FFPE tissues by RT-qPCR, and second, after the determination for the optimal cut-off of expression, we aimed to evaluate not only the accuracy but also the concordance of our technique with the currently accepted methods (IHC and FISH).
2. Materials and Methods
ACCEPTED MANUSCRIPT
2.1. Samples collection:
Formalin-fixed, paraffin-embedded (FFPE) tumor samples were retrospectively collected from 192 breast cancer patients obtained from the pathology Department of the Centre Hospitalier Universitaire (CHU) (Casablanca, Morocco) and the pathology department of the hospital clinic of Barcelona (Barcelona, Spain) . The patients were accrued consecutively and the criterion for inclusion in the study was her2 status known and none of them had received radiotherapy or chemotherapy prior to surgery. The samples were routine surgical specimens, from segmental resection or mastectomy, which had been fixed in formalin, processed, and stored according to standard histological protocols. All patients were advised of the procedures and consent. The study protocol was approved by The Ethics Committee of the school of medicine in Rabat.
2.2. Cell lines and cell culture:
Human breast carcinoma cell lines SK-BR3 (ATCC® HTB-30™) that over-expressed HER2 and MCF7 with normal-expressing HER2 (ATCC® HTB-22™) were purchased from the American Type Culture collection (ATCC, Manassas, VA, USA). The two cell lines were cultured in ATCC-formulated McCoy's 5a Medium Modified and ATCC-formulated Eagle's Minimum Essential Medium, respectively, supplemented with 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin antibiotic solution following the manufacturer instructions (ATCC, Manassas, VA, USA). All cells were maintained at 37°C, in a humidified incubator of 5% CO2, and passages upon reaching ~80% confluence.
2.3. Identification of optimal RGs
Seven genes were selected (GAPDH, PPIA, MRPL19, PUM1, UBB, RPLP37, and RPL30) to identify the most stably-expressed ECG to be used in RT-qPCR studies of HER2 gene expression in breast cancer. The ECG candidates were selected based on previous reports (Livak and Schmittgen, 2001; Drury et al., 2009; Kılıç et al., 2014).As far as we know, all these genes are expressed constitutively in breast tissues and have unrelated cellular functions and are not co-regulated. In order to identify the best ECG for breast cancer gene expression quantification, RT-qPCR was used for the generation of expression data. The obtained Ct values were used in GenEx V6 Standard software (bioMCC, Freising, Germany) that combines two statistical analysis tools geNorm and Normfinder.
GeNorm algorithm calculate stability of tested ECGs, this stability is reflected by M-value.
Progressively, the ECG gene that shows the highest variation relative to all other genes was
ACCEPTED MANUSCRIPT
eliminated. Genes with the lowest M values are the most stably expressed, while the highest M values indicate the least stable expression.
NormFinder is based on an ANOVA mathematical model and calculates a global average expression of all the genes in all the samples, to which the individual genes are compared.
Based on this comparison, SD for each candidate ECG is estimated. A SD value indicates a more stably expressed gene. NormFinder in GenExalso calculates the accumulated SD (Acc.S.D.),which indicates the optimal number of ECGs if multiple ECGs are used for normalization. A minimum value in the Acc SD corresponds to the optimum number of ECGs for accurate and reliable normalization (Andersen et al., 2004).
2.4. Immunohistochemistry IHC
Five micrometer sections of FFPE tumors were used to determine the expression of Her2 protein. The immunohistochemical test was performed in the anatomopathological laboratory of the CHU using HERcepTest and following the instructions of the manufacturer (DAKO, Glostrup, Danmark).
The samples were scored as 0, 1+, 2+ or 3+, where 0 and 1+ are HER2- and 3+ are HER2+.The samples scored as 2+ or as equivocal (weakly positive) were further confirmed by fluorescence in situ hybridization (FISH) method as recommended in guidelines.
2.5. Fluorescence in situ hybridization FISH
The 61 samples scored as 2+ were confirmed by FISH method. The FISH was performed on 3 µm paraffin tissue sections using the HER2 FISH pharmDx™ (Dako Denmark A/S, Glostrup, Denmark). The results evaluated using a fluorescence microscope equipped with appropriate filter according to the manufacturer’s recommendations.
2.6. RNA Extraction and cDNA synthesis:
Total RNA was extracted from FFPE samples (three 10-μm sections) using the Qiagen RNeasy Mini Kit extraction procedure (Qiagen, Crawley, England) and PureLink™ FFPE RNA Isolation Kit (Invitrogen, Carlsbad, Californie) according to the manufacturer’s instructions. Total RNA was eluted in 50 μL of RNase-free water and stored at –80°C until required.
Total RNAs were evaluated quantitatively and qualitatively using a NanoDrop ND-2000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA) and a Bioanalyzer (BioRad, CA, USA) respectively.
ACCEPTED MANUSCRIPT
The reaction was performed in a total volume of 20 µL using the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA), and up to 2 µg RNA was used for the reverse transcription (RT) reaction following the manufacturer’s instructions.
TaqMan RT PCRs were performed on a Veriti (Applied Biosystems, Foster City, CA, USA) following thermocycling conditions: 25°C for 10 min, 2 incubations of 37°C for 60 min and 84°C for 30 s. The cDNA was stored at −20°C.
2.7. Reverse transcription-quantitative realtime -PCR:
2.7.1. Primers and probes design
The primers and probes sequences, that were designed using Primer 3 software version 2.0, are shown in table.1. The probes were synthesized by Eurofins (Germany) with a 5’6- carboxyfluorescein (FAM) for both HER2 and the control genes as reporter dyes, as well as a 3’Black Hole Quencher-1 (BHQ-1) as a quencher dye for both probes. The primer design excluded amplification of genomic DNA. Specificity validation of the primers and probes used in RT-qPCR was based on the quantification of HER2 mRNA levels in 2 cell lines, MCF-7 and SKBR3.
2.7.2. RT-qPCR
HER2 mRNA levels were measured by RT-qPCR based on the TaqMan method in a QuantStudio6 FLEX PCR Detection System (Applied Biosystems, Foster City, CA). RT- qPCR data analysis was performed with QuantStudio 6 Flex software V1.0.The results are expressed as relative levels of HER2 mRNA referred to a calibrator sample, the MCF-7 cell line, chosen to represent 1× expression of this gene. SKBR-3 cell lines were used as a positive control since they over express the her2 gene. All analyzed tumors expressed n-fold HER2 mRNA relative to the calibrator.
The RT-qPCR mixes were prepared according to TaqMan Fast Universal PCR Master Mix instructions (Applied Biosystems, Foster City, CA). The PCR mix contained 12.5 µL Universal Master Mix, 300 nM both primers, and 200 nM probe in a total volume of 25 µL. A total of 5 µL cDNA (100 ng) was added to 20 µL PCR Mix. The reactions were performed in duplicate.The PCR cycler conditions were the same for the 3 transcripts: 95°C for 20 s, 50 cycles at 95°C for 1 s, and a combined annealing and extension step at 60°C for 30 s.
The most stable ECGs determined in this study have been used for a relative quantification of HER2 protein, fold change estimates were calculated using the geometric mean of ECG
ACCEPTED MANUSCRIPT
quantities relative to the calibrator sample and the errors were calculated following the rules of error propagation described previously (Ross and Fletcher, 1999).
2.7.3. Relative Quantity = 2–ΔΔCt
The ΔCt is calculated by normalizing the Ct of the target sample with the Ct of the endogenous control (Ct target – Ct endogenous control). The ΔΔCt then is calculated by subtracting the average ΔCt for the calibrator sample from the corresponding average ΔCt for the target sample. The relative levels of the target protein expression are expressed as a fold change relative to the calibrator sample. A relative quantity of 1 indicates no change in expression levels.
The cut-off ratio to assign negative and positive status for our method was established using univariate partition method (XLSTAT software). The RT-qPCRdata were classified into two expression levels: positive (overexpression) and negative (normal). Final result was expressed as a normalized ratio considered as over-expressed if >8.04.
2.8. Statistical analysis
The stability of th ECGs was calculated with geNorm software. Comparisons between HER-2 expression between RT-qPCR, IHC and FISH was done using Excel Stat. Receiver operating characteristic (ROC) curve analysis between IHC, FISH and HER2 mRNA RT-qPCRwas performed for all data was done with the online program http://forge.info.univ-angers.fr/
3. Results
3.1. Identification of the most stable ECG human breast cancer cell lines grown jn the laboratory
In order to identify the most suitable ECG for relative quantification of HER2 gene expression in breast cancer cells, a panel of seven gene candidates, commonly utilized in different cellular context as ECG for RT-qPCR based quantification , has been selected. This panel includes GAPDH, PPIA, MRPL19, PUM1, UBB, RPLP37, and RPL30 all of which have been already used, according to the literature, for the quantification of different genes by RT-qPCR (Drury et al., 2009; Kılıç et al., 2014). The expression analysis of these genes were first performed using well-established human breast cancer cell lines used routinely in our laboratory including MCF7 cell line which is known to express HER2 protein at a normal levels and SKBR3 that is overexpressing HER2 protein (Ross and Fletcher, 1998). Different cellular RNA preparations from the same cell line have been used. Fig. 1 summarizes the
ACCEPTED MANUSCRIPT
mean Ct value obtained in each case. Despite the use of the same amount of cDNA, an important fluctuation was observed in Ct values between different tissues for certain analyzed genes, such as GAPDH and PPIA indicating likely an unstable expression of these genes at least in the studied cell samples (Fig. 1). However, a fairly stable Ct values is obtained for mainly RPL30, RPL37 and MRPL19 genes suggesting that these genes are stably expressed throughout the analyzed cell lines (Fig. 1a and 1b). Nevertheless, the analysis of the expression stability of these genes could still not be accurately achieved based on their Ct values.. Therefore, we used the well-established GenEx V6 Standard software (bioMCC, Freising, Germany) which combines two statistical analysis tools geNorm and NormFinder as described in the material and methods. According to geNorm analysis and as indicated in Figure 2a the most stable genes with the lower M value were MRPL19, RPL30 and RPL37.
These are well known constitutively expressed genes that encode for mitochondrial ribosomal protein L19, ribosomal protein 30 and ribosomal protein 37a respectively. In line with this, NormFinder analysis has also identified MRPL19, RPL30 and RPL37 as the most stably expressed genes (Table.2) which indicate a high concordance between the two different softwares demonstrating the accuracy of the obtained results.
To extend our analysis for the determination of the most stable ECG, the same analysis described above was performed using this time human breast cancer FFPE biopsies. As described in figure. 1c and d, a fluctuation is observed in the Ct values obtained for the same gene for each analyzed biopsy sample. However, it is still difficult to determine which gene is the best ECG by using only the CT value. Therefore, we used the geNorm software to measure the stability of these genes in the used biopsies. As indicated in figure 2a, MRPL19, RPL30 and RPL37 are the genes with the lowest M value are indicating therefore that these are the most stable genes and confirming the results obtained for the breast cancer cell lines described above (Fig. 2c). To strengthen our results, we use Normfinder program and calculate the stability of the analyzed genes. Table 3 describes the stability factor obtained for each gene in the different analyzed biopsy samples. Once again, the most stable genes found by Normfinder software are MRPL19, RPL37 and RPL30 which have the lowest stability value (Table.3). This confirms the results obtained by Genorm program (Figure 2).
Furthermore, another parameter that should be taken into consideration for the selection of the best gene candidate is the expression level of this gene. In fact, the selected gene has to be expressed with high level in order to be easily detected in a given samples (Vandesompele et al., 2002). In Figure 1b, we compare the Ct value obtained for the most stable genes identified
ACCEPTED MANUSCRIPT
here for different samples used. Ct values are conversely correlated to the amount of the DNA amplified, the lower is the Ct value the more abundant is the gene transcripts. As indicated in Figure 1b, the most abundant transcripts are RPL30 followed by RPL37 and then MRPL19.
Therefore the most two suitable ECGs are RPL37 and RPL30.
Since the level of expression of MRPL19 gene was the lowest in comparison to both RPL30 and RPL37, and in addition, the efficiency of the amplification reaction for both RPL30 and RPL37 was close to the amplification efficiency of HER2 (data not shown), only RPL37 and RPL30 will be used for the normalization of the HER2 quantity in further experiments. In addition, since it is well established that normalization using multiple ECGs gives more accurate quantification then using one ECG (Bustin et al., 2005), we went to use the geometric mean for an accurate averaging of the control genes.
3.2. Cut-off determination:
ROC analysis was performed for 100 samples to determine the optimal diagnostic cut-off value with minimized false-negative and false-positive. ROC analysis was performed as described in material and methods and both HER2 mRNA levels obtained with RT-qPCR analyses and HER2 IHC score were used. The 100 FFPE samples were divided into two groups: 50 HER2 IHC-positive samples and 50 HER2 IHC negative samples.
According to the ROC analysis, the optimal cut-off value in our context is the value of 8.04 corresponding to the combination of the best sensitivity and specificity (94,59% and 76,74%).
The area under curve which is an important parameter that determines the diagnostic value of the test was also calculated (AUC = 0.92). A marker with a strong discriminative power must have an AUC value close to 1 and far from 0.5 which is the case in our study.
Based on this result, samples with a relative HER2 mRNA expression level greater than 8.04 by HER2 RT-qPCR assay were considered positive in this study. (Figure. 3)
3.3. HER2 mRNA expression in the clinical validation:
After the determination of the cut-off value a validation of our RT-qPCR-based test was performed using 116 well scored samples from the hospital clinic of Barcelona. This validation was carried out in the Pathology Department in hospital clinic of Barcelona which represents an international reference center. HER-2 status in all histological samples was assessed using FISH and IHC. Total RNA was successfully extracted from all the 116 FFPE
ACCEPTED MANUSCRIPT
breast cancer tissues. Transcript expression was evaluated by RT-qPCR in FFPE tissues.
HER-2 data obtained in all the procedures are shown on Figure 4
Concerning the IHC and RT-qPCR all the samples determined to be positive by IHC were also positive by RT-qPCR whereas 96% of samples determined to be negative by IHC were negative by RT-qPCR giving rise to a perfect overall concordance between the two tests that reached 98.18% (Table.4)
All the 61 equivocal cases (IHC 2+) were subjected to FISH analysis and there were 85.24%
cases IHC 2+/FISH - (52/61) and 14.75% IHC 2+/FISH + (9/61) (Table .4)
Concordance between FISH and RT-qPCR was also good in FISH non amplified cases with a value of 75% (39 negative RT-qPCR/ 52 FISH non amplified cases) whereas it’s 55.55 for FISH amplified case (5 positive RT-qPCR/ 9 FISH amplified case) (Table.4). The overall concordance between the two tests was 72%.
4. DISCUSSION
Accurate assessment of HER2 status is a prerequisite step for an adequate management and treatment of breast cancer patients. HER2 measurement can be performed at the DNA, mRNA, or protein level using several different methods (Ross and Fletcher, 1998). One of the most prominent methods used for the quantification of gene expression is the reverse transcription-quantitative real-time polymerase chain reaction (RT-qPCR) that constitutes a valuable tool for the retrospective analysis of clinical samples. Nevertheless, the relatively degraded nature of RNA isolated from formalin-fixed paraffin embedded (FFPE) tissue can lead to minor variations between samples in the detection of gene expression, hence, it is compulsory to select the most stable ECGs that can counter this variation and make the HER2 mRNA relative quantification more accurate. In this study we assessed the ability of a number of genes (seven genes) to be suitable references for the specific normalization of HER2 gene transcripts in FFPE breast cancer biopsies. These genes were selected from the literature and are commonly used as ECGs in many RT-qPCR based gene expression quantification studies.
We first analyzed their expression stability in several human breast cancer cell lines used routinely in our laboratory such as MCF7 and MDA231 cell lines which express HER2 protein at a normal level and SKBR3 cell line that overexpresses HER2 protein (Ross and Fletcher, 1999). Two comparative ΔCt methods, GeNorm and NormFinder have been used for this analysis and both have identified the three housekeeping genes MRPL19, RPL30 and RPL37 as the most stably expressed genes between the different analyzed cell lines. These are
ACCEPTED MANUSCRIPT
well known constitutively expressed genes that encode for mitochondrial ribosomal protein L19, ribosomal protein 30 and ribosomal protein 37a respectively. In addition to their stable expression and according to their Ct values the expression level of these genes was fairly enough to be easily detectable. Indeed, the expression level is another factor required for a gene to be selected as the best ECGs for a relative quantification (Vandesompele et al., 2002).
Using the same analysis methods, the results obtained in breast cancer cell lines were confirmed in FFPE biopsy samples. Indeed, RPL30 and RPL37 were also identified as the most stably expressed in FFPE samples. This highly suggests that these two genes are the best reference genes for an accurate relative quantification of HER2 gene expression in breast cancer.
Following the determination of RPL30 and RPL37 as the most stable ECGs, we use these two genes to accurately quantify the expression level of HER2 transcripts in different biopsy samples. In these samples, HER2 statue has been already determined by IHC and confirmed by FISH and treatment has been already given based on the outcome of these two methods.
We first went to determine the optimal diagnostic cut-off using ROC-based analysis of 100 biopsy samples with HER2 score already determined by IHC. The ROC analysis is considered as the most efficient statistical method for cut-off determination in biological system (Hernández-Orallo et al., 2013). Indeed, a cut-off of 8.04 was determined as the optimal value that represents the best specificity and sensitivity combination. This cut-off value represents the amount of HER2 transcripts relative to the ECGs transcripts above which HER2 is considered to be overexpressed in our RT-qPCR based test. It is noteworthy to mention that determination of an accurate cut-off value has always been a challenge for gene expression quantification in general and particularly in the HER2 testing. Indeed, a threshold of gene expression is one limiting factor for an accurate HER2 quantification in breast cancer and therefore treatment orientation. Using the determined cut-off value, we analysed the validity of our RT-qPCR developed test by measuring the level of expression of HER2 transcripts in 116 new breast cancer biopsy samples in the Pathology Department of hospital clinic of Barcelona which is an international reference center in molecular pathology testing validation. 55 of 116 samples were scored positive or negative and 61 were scored equivocal by IHC and confirmed by FISH. HER2 quantification was performed using the geometric average of the two ECGs for the normalisation. Our RT-qPCR-based test was very highly concordant with IHC reaching a value of 98% that is above the rate of concordance recommended by the American Society of Clinical Oncology and the College of American
ACCEPTED MANUSCRIPT
Pathologists (ASCO/CAP 2013 guidelines) (Wolff et al., 2007). Moreover, the concordance of our test with FISH analysis was also good reaching a value of 72%. Although, this concordance was not as high as with IHC, this is somehow expected since it is well established that FISH and IHC are not always highly concordant and up to 20% of interlaboratory discordance rates have been described in several studies (Vinatzer et al., 2005;
Baehner et al., 2010; Lehmann-Che et al., 2011). This makes it difficult for a test to be highly concordant with two tests in the same time. Nevertheless, further studies with larger sample size in trastuzumab-treated patients are still recommended to find out whether RT-qPCR based quantification of HER2 expression has advanced benefits to choose anti-HER2 therapeutics or other therapies.
Adaptation of both RNA purification and reverse transcription procedures has facilitated the use of RT-qPCR to study gene expression at the level of mRNA (Bustin et al., 2005).
Nevertheless, getting reliable and accurate quantification using RT-qPCR requires suitable normalization to correct for non-biological variables such as starting template amounts, RNA quality and both reverse transcription and PCR reactions efficiency. Therefore, the most commonly used method remains the use of endogenous control genes (ECG). The issue of carefully selecting and validating ECG has been a matter of discussion for many experimental systems in the context of RT-qPCR for mRNA (Tricarico et al., 2002; Bustin et al., 2005).
However, this matter has not yet been carefully explored in relation to the relative quantification of HER2 protein expression in breast tissue. This study constitutes, to the best of our knowledge, the first to identify genes encoding for ribosomal proteins (RPL30 and RPL37) as the best ECG for HER2 quantification. Furthermore, this is the first systematic evaluation study of ECG candidates for the specific normalization of HER2 gene expression using RT-qPCR in breast cancer. Indeed, previous studies describing the identification of new ECGs for RT-qPCR were all focusing on the relative quantification of different other genes than HER2 in different tissue samples including breast cancer (BOSSARD et al., 2005). In addition, in many studies describing RT-qPCR-based quantification of HER2 mRNA, different well-established housekeeping genes have been used as ECG assuming that they are stably expressed between the different studied tissues. For these reason most of studies assessing the accuracy of RT-qPCR in HER2 gene expression status determination using such reference genes have shown contradictory results. This was mainly due to the lack of reliable quantification normalization by a suitable ECG which hugely accounts for a consistent quantification. .
ACCEPTED MANUSCRIPT
5. CONCLUSION
In conclusion, RT-qPCR could be an easy, accurate and complementary method to IHC for the measurement of HER2 gene expression when the design criteria of carefully selected ECGs and short amplicons are used. This will enable more efficient use of archives FFPE, the most abundant tissue resource available to most institutions. Although the main objective of the current study is to measure the concordance of RT-qPCR based quantification of HER2 overexperession with both IHC and FISH, it is still interesting to find out whether RT-qPCR based quantification of HER2 expression has advanced benefits to choose anti-HER2 therapeutics or other therapies (Yamamoto-Ibusuki et al., 2013; Yamamoto-Ibusuki et al., 2015). This will be the scope of a new study currently performed in our laboratory.
ABREVIATIONS
RT-qPCR: reverse transcription quantitative reel time polymerase chain reaction HER2: human epidermal growth factor receptor 2
GAPDH: Glucophosphate dehydrogenase PPIA: Peptidylprolyl isomerase A
RPL37a: Ribosomal protein L37-alpha RPL30: Ribosomal protein L30
MRPL19: mitochondrial ribosomal protein L19 PUM1: pumilio RNA-binding family member 1 UBB: ubiquitin B
IHC: immunohistochemistry
FISH: fluorescence in situ hybridization, ECG: endogenous control gene
ETHICS APPROVAL AND CONSENT TO PARTICIPATE.
Ethical approval for the use of patient’s samples has been obtained from the Rabat medical school ethic committee
COMPETING INTERESTS
The authors declare that they have no competing interests.
ACCEPTED MANUSCRIPT
FUNDING
This work was funded by both the Moroccan foundation for advanced science, innovation and research (MAScIR) and the OCP company.
AUTHOR’S CONTRIBUTIONS
HE carried out most of the experiments. IAM, ZQ, DK, ME and NB helped in performing experiments. HE, IAM, ZQ, DK, ME and NB participated in the design of the study, performed the statistical analysis and helped in the drafted the manuscript. BY, HS participated in the design of the study and were involved with revising the manuscript critically. MK, HA and HE provided and managed the patient’s samples and revised the manuscript critically.PF conducted the external validation of the test in Barcelona. AM conceived, designed, coordinated the study and drafted the paper. All authors read and approved the final manuscript.
ACKNOWLEGEMNTS
We thank all patients that have kindly agreed to provide us with the samples analyzed here.
We are also very grateful to all members of the medical biotechnology laboratory in MAScIR for their help in performing this research. We are thankful to the technical staff of Pathology Department in Hopital Ibn Rochd Casablanca and National Institute of Oncology Rabat.We are indebted to MAScIR administration staff for their support.
ACCEPTED MANUSCRIPT
Table.1
Sequence of the primers and probes used to amplify HER2, RPL30 and RPL37A genes
Her2 (Human epidermal growth factor receptor 2)
Her2 (forward) 5’-AATGCCAGGCACTGTTTG-3’
Her2 (reverse) 5’-GTCCTTATAGTGGGCACAGG-3’
Her2 (probe) 5’- (FAM)CCGTGCCACCCTGAGTGTCA(TAMRA)-3’
RPLP37A (Ribosomal protein 37a)
PRPL37A (forward) 5’-ACATGGCCAAACGTACCA-3’
PRPL37A (reverse) 5’-TGCTGGCTGATTTCAATTTT-3’
PRPL37A (probe) 5’-(FAM)CGGTAAATACGGGACCCGCTATG(TAMRA)-3’
RPL30 (RPL30 ribosomal protein 30)
RPL30 (forward) 5’-ATGGCCAAACGTACCAAG-3’
RPL30 (reverse) 5’-AAGTGTACTTGGCGTGCTG-3’
RPL30 (probe) 5’-(VIC)TTCACCAGTCTGTTCTGGCATGC(TAMRA)-3’
Table.2
Expression stability values of the candidate ECGs calculated using cell line by Normfinder Gene name Stability value
GAPDH 0.275
PPIA 0.235
MRPL19 0.064
PUM1 0.22
UBB 0.097
RPLP37 0.032
RPL30 0.074
ACCEPTED MANUSCRIPT
Table.3
Expression stability values of the candidate ECGs calculated using biopsy samples by Normfinder
Gene name Stability value
GAPDH 1.036
PPIA 0.578
MRPLP19 0.171
PUM1 0.768
RPLP37 0.238
UBB 0.499
RPL30 0.491
Table.4
Concordance results of MASCIR Q-RT-PCR test with in routine diagnostic HER2 assays
Routine Test Test results MASCIR results
Concordance (%) (QRTPCR)
IHC
0/1+
n=33
Negn=32
96,96%
Pos n=1
3+
n=22
Negn=0
100,00%
Pos n=22
Total n=55 n=55 98,18
FISH NA n=52
Neg
n=39 75%
ACCEPTED MANUSCRIPT
Pos n=13
A n=9
Neg
n=4 55,55%
Pos n=5
Total n=61 n=61 72,13%
N : patients number NA: Non amplified - A : amplified - Neg : negative - Pos : positive
ACCEPTED MANUSCRIPT
REFERENCES
Andersen, C.L., Jensen, J.L. and Ørntoft, T.F., 2004. Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited
for normalization, applied to bladder and colon cancer data sets. Cancer research 64, 5245- 5250.
Baehner, F.L., Achacoso, N., Maddala, T., Shak, S., Quesenberry, C.P., Goldstein, L.C., Gown, A.M. and Habel, L.A., 2010. Human epidermal growth factor receptor 2 assessment in a case-control study: comparison of fluorescence in situ hybridization and quantitative reverse transcription polymerase chain reaction performed by central laboratories. Journal of Clinical Oncology 28,
4300-4306.
Barberis, M., Pellegrini, C., Cannone, M., Arizzi, C., Coggi, G. and Bosari, S., 2008. Quantitative PCR and HER2 testing in breast cancer. American journal of clinical pathology 129, 563-570.
Baselga, J., Gianni, L., Geyer, C., Perez, E.A., Riva, A. and Jackisch, C., 2004. Future options with trastuzumab for primary systemic and adjuvant therapy, Seminars in oncology. Elsevier, pp.
51-57.
Bast, R.C., Ravdin, P., Hayes, D.F., Bates, S., Fritsche, H., Jessup, J.M., Kemeny, N., Locker, G.Y., Mennel, R.G. and Somerfield, M.R., 2001. 2000 update of recommendations for the use of tumor markers in breast and colorectal cancer: clinical practice guidelines of the American
Society of Clinical Oncology. Journal of Clinical Oncology 19, 1865-1878.
BOSSARD, C., BIECHE, I., LE DOUSSAL, V., LIDEREAU, R. and SABOURIN, J.C., 2005. Real-time RT-PCR: a complementary method to detect HER-2 status in breast carcinoma. Anticancer research 25,
4679-4683.
Bustin, S., Benes, V., Nolan, T. and Pfaffl, M., 2005. Quantitative real-time RT-PCR–a perspective.
Journal of molecular endocrinology 34, 597-601.
Callahan, R., 1989. Genetic alterations in primary breast cancer. Breast cancer research and treatment 13, 191-203.
Chang, T., Juan, C., Yin, P., Chi, C. and Tsay, H., 1998. Up-regulation of beta-actin, cyclophilin and GAPDH in N1S1 rat hepatoma. Oncology reports 5, 469-540.
Drury, S., Anderson, H. and Dowsett, M., 2009. Selection of REFERENCE genes for normalization of qRT-PCR data derived from FFPE breast tumors. Diagnostic Molecular Pathology 18, 103-107.
Esteva, F.J., Sahin, A.A., Cristofanilli, M., Coombes, K., Lee, S.-J., Baker, J., Cronin, M., Walker, M., Watson, D. and Shak, S., 2005. Prognostic role of a multigene reverse transcriptase-PCR assay in patients with node-negative breast cancer not receiving adjuvant systemic therapy. Clinical
cancer research 11, 3315-3319.
Gutierrez, C. and Schiff, R., 2011. HER2: biology, detection, and clinical implications. Archives of pathology & laboratory medicine 135, 55-62.
Hanna, W.M., Rüschoff, J., Bilous, M., Coudry, R.A., Dowsett, M., Osamura, R.Y., Penault-Llorca, F., Van De Vijver, M. and Viale, G., 2014. HER2 in situ hybridization in breast cancer: clinical
implications of polysomy 17 and genetic heterogeneity. Modern Pathology 27, 4-18.
Hernández-Orallo, J., Flach, P. and Ferri, C., 2013. ROC curves in cost space. Machine Learning 93, 71- 91.
Kılıç, Y., Celebiler, A. and Sakızlı, M., 2014. Selecting housekeeping genes as references for the normalization of quantitative PCR data in breast cancer. Clinical and Translational Oncology
16, 184-190.
Lehmann-Che, J., Amira-Bouhidel, F., Turpin, E., Antoine, M., Soliman, H., Legres, L., Bocquet, C., Bernoud, R., Flandre, E. and Varna, M., 2011. Immunohistochemical and molecular analyses
ACCEPTED MANUSCRIPT
of HER2 status in breast cancers are highly concordant and complementary approaches.
British journal of cancer 104, 1739-1746.
Livak, K.J. and Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2− ΔΔCT method. methods 25, 402-408.
Pazhoomand, R., Keyhani, E., Banan, M., Najmabadi, H., Khodadadi, F., Iraniparast, A., Feiz, F., Majidzadeh, K., Bahman, I. and Moghadam, F.A., 2013. Detection of HER2 status in breast cancer: comparison of current methods with MLPA and real-time RT-PCR. Asian Pac J Cancer
Prev 14, 7621-8.
Pegram, M.D., Pauletti, G. and Slamon, D.J., 1998. HER-2/neu as a predictive marker of response to breast cancer therapy. Breast cancer research and treatment 52, 65-77.
Perez, E.A., Baehner, F.L., Butler, S.M., Thompson, E.A., Dueck, A.C., Jamshidian, F., Cherbavaz, D., Yoshizawa, C., Shak, S. and Kaufman, P.A., 2015. The relationship between quantitative
human epidermal growth factor receptor 2 gene expression by the 21-gene reverse transcriptase polymerase chain reaction assay and adjuvant trastuzumab benefit in Alliance
N9831. Breast Cancer Research 17, 1.
Press, M.F., Slamon, D.J., Flom, K.J., Park, J., Zhou, J.-Y. and Bernstein, L., 2002. Evaluation of HER- 2/neu gene amplification and overexpression: comparison of frequently used assay methods
in a molecularly characterized cohort of breast cancer specimens. Journal of Clinical Oncology 20, 3095-3105.
Rondinelli, R., Epner, D. and Tricoli, J., 1997. Increased glyceraldehyde-3-phosphate dehydrogenase gene expression in late pathological stage human prostate cancer. Prostate cancer and
prostatic diseases 1, 66-72.
Ross, J.S. and Fletcher, J., 1999. HER-2/neu (c-erb-B2) gene and protein in breast cancer. American journal of clinical pathology 112, S53-67.
Ross, J.S. and Fletcher, J.A., 1998. The HER‐2/neu Oncogene in Breast Cancer: Prognostic Factor, Predictive Factor, and Target for Therapy. Stem cells 16, 413-428.
Sauter, G., Lee, J., Bartlett, J.M., Slamon, D.J. and Press, M.F., 2009. Guidelines for human epidermal growth factor receptor 2 testing: biologic and methodologic considerations. Journal of
Clinical Oncology 27, 1323-1333.
Slamon, D.J., Leyland-Jones, B., Shak, S., Fuchs, H., Paton, V., Bajamonde, A., Fleming, T., Eiermann, W., Wolter, J. and Pegram, M., 2001. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. New England Journal of
Medicine 344, 783-792.
Tricarico, C., Pinzani, P., Bianchi, S., Paglierani, M., Distante, V., Pazzagli, M., Bustin, S.A. and Orlando, C., 2002. Quantitative real-time reverse transcription polymerase chain reaction:
normalization to rRNA or single housekeeping genes is inappropriate for human tissue biopsies. Analytical biochemistry 309, 293-300.
Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe, A. and Speleman, F., 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging
of multiple internal control genes. Genome biology 3, 1.
Vinatzer, U., Dampier, B., Streubel, B., Pacher, M., Seewald, M.J., Stratowa, C., Kaserer, K. and Schreiber, M., 2005. Expression of HER2 and the coamplified genes GRB7 and MLN64 in
human breast cancer: quantitative real-time reverse transcription-PCR as a diagnostic alternative to immunohistochemistry and fluorescence in situ hybridization. Clinical cancer
research 11, 8348-8357.
Wolff, A.C., Hammond, M.E.H., Schwartz, J.N., Hagerty, K.L., Allred, D.C., Cote, R.J., Dowsett, M., Fitzgibbons, P.L., Hanna, W.M. and Langer, A., 2007. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. Archives of pathology &
laboratory medicine 131, 18-43.
ACCEPTED MANUSCRIPT
Yamamoto-Ibusuki, M., Yamamoto, Y., Fu, P., Yamamoto, S., Fujiwara, S., Honda, Y., Iyama, K.-i. and Iwase, H., 2015. Divisional role of quantitative HER2 testing in breast cancer. Breast Cancer
22, 161-171.
Yamamoto-Ibusuki, M., Yamamoto, Y., Yamamoto, S., Fujiwara, S., Fu, P., Honda, Y., Iyama, K.-i. and Iwase, H., 2013. Comparison of prognostic values between combined immunohistochemical score of estrogen receptor, progesterone receptor, human epidermal growth factor receptor 2, Ki-67 and the corresponding gene expression score in breast cancer. Modern Pathology 26,
79-86.
ACCEPTED MANUSCRIPT
FIGURES LEGENDES
Figure.1 Ct value distribution for the analyzed genes
(a) Total RNA was extracted from different cell lines and following reverse transcription quantitative real time PCR was performed on obtained cDNA. Variation of the Ct value obtained for the each analyzed gene is represented throughout the used cell line samples. (b) Representation of the CT value variation of the selected most stable genes in the used cell lines. (c) as in (a) Variation of the Ct value obtained for each analyzed gene is represented all used biopsy samples (d) Representation of the CT variation of the selected most stable genes in the used biopsy samples.
Figure.2 GeNorm analysis of the ECGs.
Results are presented as per the output file of the geNorm program. (a and b) Stepwise exclusion of the least stable genes in cell line (a) and biopsies (b). The gene stability value M is based on the average pairwise variation between all tested genes. Low M values indicate more stable genes. The analysis indicates that RPL37, RPL30 and RPL19 are the most stable genes.
Figure.3 Sensitivity and specificity of our test by ROC analysis
The receiver operating characteristic (ROC) curve analysis of HER2 expression in biopsies.
The ROC curve is used to determine both sensitivity and specificity of the test. The optimal cut-off value in our context is the value of 8.04 corresponding to the combination of the best sensitivity and specificity (94,59% and 76,74%) . The area under curve which is a parameter to determine the diagnostic value of the test was also calculated (AUC = 0.92)
Figure.4 Comparison of HER2 expression using RTqPCR and IHC/FISH methods The results show the comparison between RTqPCR and IHC/FISH in the discrimination of positive and negative samples. (a) Representation of the comparison of samples analyzed by RTqPCR and IHC. IHC- are the samples determined to be negative by IHC and IHC+ are the positive one. The red line represents the threshold for the RTqPCR. Above the threshold are the samples shown to be positive by the RTqPCR and below the threshold are the one which are negative. (b) As in (a) RTqPCR is compared here to FISHE. FISH- is the samples sown to be negative by FISH and FISH+ are the one positive.
ACCEPTED MANUSCRIPT
EL HADI et al. Figure.1
ACCEPTED MANUSCRIPT
EL HADI et al. Figure.2
ACCEPTED MANUSCRIPT
EL HADI et al. Figure.3
ACCEPTED MANUSCRIPT
EL HADI et al. Figure.4
a
b
ACCEPTED MANUSCRIPT
ABREVIATIONS
RT-qPCR: reverse transcription quantitative reel time polymerase chain reaction HER2: human epidermal growth factor receptor 2
GAPDH: Glucophosphate dehydrogenase PPIA: Peptidylprolyl isomerase A
RPL37a: Ribosomal protein L37-alpha RPL30: Ribosomal protein L30
MRPL19: mitochondrial ribosomal protein L19 PUM1: pumilio RNA-binding family member 1 UBB: ubiquitin B
IHC: immunohistochemistry
FISH: fluorescence in situ hybridization, ECG: endogenous control gene
ACCEPTED MANUSCRIPT
Highlights
Analysis of HER2 status in breast cancer samples.
Use of qPCR in the analysis of HER2 expression in breast cancer.
Identification of the most stable control gene for an accurate analysis of HER2 expression in biopsy samples.
Clinical validation of the RT-qPCR based HER2 quantification test.
