CLB-08858; No. of pages: 6; 4C: Clinical Biochemistry xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
BCL2L12: A promising molecular prognostic biomarker in breast cancer Alexandros Tzovaras a, Athina Kladi-Skandali b, Kleita Michaelidou b, George C. Zografos c, Ioannis Missitzis d, Alexandros Ardavanis a, Andreas Scorilas b,⁎ a
First Department of Medical Oncology, “Saint Savvas”, Anticancer Hospital, 11522, Athens, Greece Department of Biochemistry and Molecular Biology, University of Athens, 15701 Athens, Greece First Propaedeutic Surgical Department, Athens Medical School, “Hippokration” General Hospital, Athens, Greece d The Breast Clinic “Saint Savvas”, Anticancer Hospital, 11522,Athens, Greece b c
a r t i c l e
i n f o
Article history: Received 1 July 2014 Received in revised form 3 September 2014 Accepted 4 September 2014 Available online xxxx Keywords: BCL2 family Apoptosis Molecular tumor markers Prognostic biomarker Quantitative real-time PCR
a b s t r a c t Objectives: BCL2-like 12 (BCL2L12) is a new member of the BCL2 gene family that was discovered and cloned by members of our group and found to be expressed in the mammary gland. Many genes of the BCL2 family were found to be implicated in breast carcinogenesis and to serve as possible prognostic markers. The aim of the present study was the quantification of BCL2L12 mRNA expression in order to assess its value as a prognostic tissue biomarker in breast cancer (BC). Design and methods: BCL2L12 mRNA levels were determined in a statistically significant sample size of cancerous (N = 108) and adjacent non-cancerous (N = 71) breast tissues using a highly sensitive quantitative real-time polymerase chain reaction (qRT-PCR) method. Relative quantification analysis was conducted using the comparative CT (2− ΔΔCT) method, whereas the association between BCL2L12 expression and clinopathological data, disease-free survival (DFS) and overall survival (OS) were estimated by statistical analysis. Results: BCL2L12 mRNA expression was decreased in malignant samples compared to the histologically normal counterparts (p = 0.012). Significant relationships between BCL2L12 expression and TNM stages (p = 0.009), metastatic potential (p = 0.012), tumor size (p = 0.04) and age (p = 0.024) were observed. Moreover, Kaplan–Meier and Cox univariate analyses indicated that BCL2L12 expression is associated with longer DFS, whereas multivariate analysis pointed out the independent favorable prognostic value of BCL2L12. Conclusions: According to our results, BCL2L12 mRNA expression is a favorable prognostic marker of DFS for BC patients, suggesting its possible application as a novel prognostic indicator of this malignancy. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Breast cancer (BC), the leading cause of cancer death in women worldwide [1], is a multifactorial and particularly heterogeneous malignancy [2]. In the genomic era, the traditional clinical–pathological approach of BC prognosis seems inadequate to describe the disease complexity and therefore molecular biomarker tests as well as gene expression signatures constantly gain ground in BC patients' management [3,4]. Nowadays, two real-time PCR-based assays (Oncotype DX and PAM50) and a microarray-based multigene test (MammaPrint) are commercially available and enable the accurate classification of BC patients, estimation of recurrence risk and prediction of treatment
Abbreviations: BC, breast cancer; qRT-PCR, quantitative real-time polymerase chain reaction;DFS,disease-free survival; OS, overallsurvival; RQ units,relative quantification units; CT, threshold cycle; NPI, Nottingham Prognostic Index. ⁎ Corresponding author at: Department of Biochemistry and Molecular Biology, University of Athens, 15701 Panepistimiopolis, Athens, Greece. Fax: +30 210 727 4158. E-mail address: [email protected] (A. Scorilas).
response [3]. These approaches allow the simultaneous analysis of the expression patterns of a set of genes, associated with several hallmarks of BC including angiogenesis, metastasis and impairment of apoptosis [5–7]. BCL2-like 12 (BCL2L12) is a novel member of the BCL2 family of apoptosis-related genes that maps to chromosome 19q13.3 [8], a genomic locus where many cancer-related genes are located, such as the KLK gene family [9], the CEACAM/PSG cluster [10] and the gene coding for the urokinase-type plasminogen activator receptor (uPAR) [11]. The BCL2L12 mRNA expression has been studied not only in solid tumors [12–16] but also in haematological malignancies [17,18], and the results are encouraging regarding the potential use of this gene as a tumor marker. As far as BC is concerned, preliminary data generated from conventional PCR supports the notion that this apoptosis-related gene is a favorable prognostic marker respecting BC patients' DFS and OS [19,20]. In light of these data, we conducted this study aiming at the verification of the prognostic value of this gene in BC. For this purpose, a sensitive quantitative RT-PCR (qRT-PCR) methodology was developed and applied for the expression analysis of BCL2L12 in a large number of BC
http://dx.doi.org/10.1016/j.clinbiochem.2014.09.008 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
2
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
patients. To the best of our knowledge, this is the first study examining the expression of BCL2L12 gene in breast carcinoma tissues using this methodology. Materials and methods Breast tissue specimens The current study was designed according to new guidelines for reporting new tumor biomarkers [21] and was approved by the institutional review boards of the Anticancer Hospital “Saint Savvas”. Research procedures of this study comply with the ethical standards of the World Medical Association Declaration of Helsinki (version: 2008). Informed consent was obtained from patients participating in the study. The study group consisted of 179 samples (71 pairs of BC tissues and their normal counterparts and 37 additional samples of tumoral tissue), obtained from women who had undergone surgery for histologically verified primary breast carcinomas. Patients were nursed at the 1st Oncology Department of the Anticancer Hospital “Saint Savvas”, Athens, Greece and sample collection took place between 2009 and 2011. All tissue specimens were snap-frozen in liquid nitrogen and preserved at − 80 °C until use. Adequate amounts of tissue for RNA extraction and detailed clinicopathological data (Supplementary Table S1) were prerequisites for specimens' further analysis. Patient age ranged from 31 up to 87 years (mean ± SD, 59.20 ± 12.83) and tumor size varied between 0.8 and 10.5 cm (mean ± SD, 3.20 ± 1.82). The median follow-up period for DFS and OS was 32.12 and 37.51 months, respectively, with a range of 0.60–47.44 months (Supplementary Table S2). RNA extraction and first-strand cDNA synthesis Total RNA was extracted from frozen pulverized breast tissue as well as from MCF-7 cells (human breast adenocarcinoma cell line) using the guanidine isothiocyanate method (TRI Reagent®, Ambion). According to the standard protocol, the RNA pellet was diluted in RNA storage solution (Ambion) and stored immediately at −80 °C until further analysis. The concentration and purity of the RNA were determined by UVspectrophotometry, while RNA quality was verified by agarose gel electrophoresis. Thereafter, first-strand cDNA was synthesized from 2 μg of total RNA according to the previously described protocol [13]. Quantitative real-time PCR (qRT-PCR) A sensitive qRT-PCR methodology (Supplementary Fig. 1) was developed for the molecular analysis of BCL2L12, utilising the SYBR Green I chemistry. Based on NCBI information, two pairs of genespecific primers were designed for the amplification of the endogenous reference gene (HPRT1; GenBank accession number: NM_000194.2) and the target gene (BCL2L12; GenBank accession number: NM_138639.1). The sequences of the BCL2L12 qRT-PCR primers were: 5′-CCCTCGGCCTTGCTCTCT-3′ and 5′-TCCGCAGTATGGCTTCCTTC-3′, producing a 182-bp PCR amplicon. The corresponding sequences of the HPRT1 primers were 5'-TGGAAAGGGTGTTTATTCCTCAT-3' and 5'ATGTAATCCAGCAGGTCAGCAA-3' giving a 151 bp PCR product. qRT-PCR for HPRT1 and BCL2L12 cDNAs was conducted in duplicate, on the same 96-well PCR microplate in a 7500 Thermal Cycler (Applied Biosystems). The 10 μl reaction mixture contained 10 ng of cDNA, 5 μl Kapa SYBR Fast Universal 2× qPCR Master Mix (Kapa Biosystems), 0.2 μl of 50 × Rox Low passive reference dye (Kapa Biosystems), 50 nM of each BCL2L12 primer/75 nM of each HPRT1 primer and DEPC-treated water to the final volume. The PCR protocol conditions consisted of an initial 3-minute polymerase activation step at 95 °C, followed by 40 cycles of 95 °C for 15 seconds for template denaturation, and 60 °C for 1 minute for primer annealing and extension. After completion of the PCR reaction, a melt curve analysis was conducted to
distinguish the PCR products of interest from any non-specific products and/or primer–dimers through their particular melting temperatures (Tm). The Tm of the BCL2L12 and HPRT1 PCR products were 84.5 °C and 81.0 °C, respectively. The relative quantification of the expression analysis of all breast specimens was conducted using the comparative CT (2−ΔΔCT) method, and the expression levels of the gene of interest are expressed as relative quantification units (RQ units). According to this approach, ΔCT represents the difference between the threshold cycle (CT) of the target gene (BCL2L12) and the CT value of the corresponding endogenous reference gene (HPRT1) in each sample under study as follows: (ΔCT = CT BCL2L12 − CT HPRT1). The ΔΔCT value is equal to the difference between the average ΔCT value of an experimental sample and the average ΔCT of the corresponding calibrator according to the formula (ΔCT test sample − ΔCT calibrator). The application of the comparative CT method requires approximately equal amplification efficiencies for target and reference genes [22]. For this reason, data from previously described validation experiments were used [13]. In the current study, HPRT1 was used as an internal control gene for normalization purposes, since it is one of the most stable housekeeping genes and its use as a single reference gene is recommended for expression studies in several types of cancer, including BC [23]. Moreover, the human BC cell line MCF-7 served as calibrator. Statistical analyses The nonparametric Wilcoxon signed-rank test was performed in order to estimate the differences in BCL2L12 expression levels between cancerous and non-cancerous tissues. To evaluate the discriminatory value of BCL2L12 expression levels between tumoral and normal tissues, receiver operating characteristic curve (ROC) was constructed and the area under the ROC curve (AUC) was analyzed by the Hanley and McNeil method [24]. In order to examine the relation of BCL2L12 expression levels, as a continuous variable, with various qualitative clinicopathological parameters, we performed either Mann–Whitney U test or Jonckheere–Terpstra test. Since there are no established cutoffs for BCL2L12 expression, the X-Tile algorithm was used to produce an optimal bifurcating point of 0.77 RQ Units (equal to 55th percentile) [25]. According to the results, the patient cohort was classified into high and low BCL2L12 expression groups and associations between this status and other categorical clinicopathological variables were analyzed using either Fisher's exact test or Pearson chi-square test. The associations between various markers of prognosis and the relative risks for disease recurrence and death of patients were determined using univariate and multivariate Cox proportional hazard regression models. The resulting p value was evaluated by the test for trend approach. The survival analysis included also the construction of Kaplan–Meier curves, which represent the percentage probability of the DFS and OS, using as threshold the aforementioned cutoff. The p value was calculated using the long-rank test. Finally, the level of significance was defined at a probability value of less than 0.05 (p b 0.05). Results Evaluation of the discriminatory potential of BCL2L12 mRNA expression in BC patients According to the descriptive statistical analysis, BCL2L12 expression was remarkably elevated in non-cancerous tissues compared to their malignant counterparts. In particular, BCL2L12 mRNA expression in the BC samples ranged from 0.02 up to 28.25 RQ units with a mean ± SD of 1.63 ± 3.83 RQ units and a median of 0.60 RQ units. On the other hand, BCL2L12 mRNA expression in the adjacent histologically normal tissues varied between 0.02 and 70.00 RQ units with a mean ± SD of
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
5.00 ± 10.00 RQ units and a median of 0.91 RQ units (Supplementary Table 2). The comparison of BCL2L12 mRNA levels in 71 pairs of breast tissues (normal versus cancerous tissue from the same patient) revealed a significantly higher expression in nonmalignant samples than in their matched tumoral tissues (p = 0.012). According to Fig. 1, 29 patients (40.8%) demonstrated increased expression levels in the malignant breast tissue compared to the normal counterpart, whereas in 39 women (55.0%), BCL2L12 expression in the cancerous specimen was lower than in the paired non-cancerous tissue. In order to assess the diagnostic value of BCL2L12 expression, we performed receiver operating characteristic (ROC) curve analysis (Supplementary Fig. S2). BCL212 could distinguish patients with BC from healthy controls, exhibiting, however, a moderate differential diagnostic performance with an AUC of 0.65 (AUC = 0.65; 95% CI = 0.56–0.74; p = 0.001). Correlations between BCL2L12 mRNA expression and BC patients' clinicopathological parameters Associations between BCL2L12 expression and patients' clinicopathological parameters such as tumor size and histological grade, molecular subtype, stage and metastasis were examined (Tables 1 and 2), revealing that BCL2L12 overexpression was found to be associated with favorable clinicopathological features. Women with early TNM stage (I/II) BC demonstrated significantly (p = 0.007) higher BCL2L12 mRNA levels (median = 0.84 RQ units) in comparison to those with stage III/IV BC (median = 0.41 RQ units; Fig. 2A). Moreover, an inverse relationship between BCL2L12 expression levels and tumor size was observed (p = 0.010). Particularly, median BCL2L12 expression in the group of patients with tumor size ≤2 cm was 1.01 RQ units, whereas in tumors of 2–5 cm and N5 cm, the median values were 0.75 and 0.22 RQ units, respectively (Fig. 2B). Finally, women in the age range of 35 to 55 years exhibited slightly higher, although not significant (p = 0.07), BCL2L12 expression (median = 1.02 RQ units) than patients aged less than 35 years (median = 0.46 RQ units) and those aged over 55 years (median = 0.48 RQ units; Table 1). Similar results were obtained when we classified patients into two categories (BCL2L12-high and BCL2L12-low) on the basis of BCL2L12 mRNA levels, and we examined associations between this status and the abovementioned categorical variables (Table 2). For this reason, a cutoff value of 0.77 RQ units (equal to the 55th percentile) was generated and the associations were examined using the Pearson chi-square and Fisher exact tests, wherever appropriate. According to these results, BCL2L12 overexpression was significantly associated with early TNM
Fig. 1. Comparison of BCL2L12 mRNA levels in malignant and histologically normal counterparts. Bar graph representing the difference in the BCL2L12 expression between cancerous and non-cancerous breast specimens. The p value was calculated using “Wilcoxon signed-rank test”.
3
Table 1 BCL2L12 mRNA expression in BC patients according to classic clinicopathological data. Variable
No. of patients
Mean ± SD
Median
p value
Histological gradea I II III
3 57 41
3.43 ± 5.40 1.74 ± 4.30 1.56 ± 3.41
0.42 0.58 0.77
0.17c
Stageb I/II III/IV
69 30
2.18 ± 4.68 0.72 ± 0.88
0.84 0.41
0.007d
Tumor size ≤2 cm 2–5 cm N5 cm
20 64 13
2.39 ± 5.10 1.84 ± 4.05 0.41 ± 0.41
1.01 0.75 0.22
0.010c
Molecular subtype Luminal A Luminal B Triple negative HER-2 group
32 27 16 8
2.15 1.56 1.03 1.84
5.43 4.18 0.72 2.55
0.59 0.52 0.85 0.65
0.24c
Type NST Lobular
74 13
1.20 ± 2.56 1.97 ± 4.47
0.42 0.75
0.20d
Age ≤35 35–55 N55
3 36 66
0.50 ± 0.40 2.47 ± 5.12 1.28 ± 3.07
0.46 1.02 0.48
0.07c
± ± ± ±
Abbreviations: SD = standard deviation. Statistically significant results are shown in bold type. a Bloom–Scarff–Richardson grading system. b TNM staging system. c Calculated by the “Jonckheere–Terpstra test”. d Calculated by the “Mann–Whitney U test”.
stages (p = 0.009), small tumor size (p = 0.04) and age range of 35–55 years (p = 0.024). Regarding BCL2L12 expression and the presence of metastases, a statistically significant negative association was demonstrated (p = 0.012). BCL2L12 mRNA status and survival of BC patients Complete follow-up data were available for 101 patients, among whom 16 (15.8%) relapsed and 7 (6.9%) died. The prognostic significance of BCL2L12 expression in our cohort of BC patients was estimated by the Kaplan–Meier survival analysis (Fig. 3) and the Cox regression (Tables 3 and 4), after classification of patients into two categories, based on BCL2L12 expression as described in the “Statistical analysis” section. According to Kaplan–Meir analysis, patients overexpressing BCL2L12 have significantly superior DFS (p = 0.015) and OS (p = 0.025; median DFS and OS time = 37.5 and 38.2 months, respectively) compared to those with decreased BCL2L12 mRNA levels (median DFS and OS time = 30.6 and 34.7 months, respectively). Cox univariate regression analysis (Table 3), also, revealed the favorable prognostic value of BCL2L12 overexpression in terms of DFS (p = 0.030), but not of OS (data not shown), as women demonstrating increased BCL2L12 expression had a significantly decreased risk of relapse (hazard ratio [HR] = 0.19, 95% confidence interval [95% CI] = 0.44–0.85; p = 0.030). In the Cox multivariate analysis (Table 4), the full multiparametric model was adjusted for tumor size, histological grade, patients' age and Nottingham prognostic index and revealed that BCL2L12 mRNA expression (hazard ratio [HR] = 0.12; 95% confidence interval [95% CI] = 0.015–0.98; p = 0.048) can predict BC patients' relapse independently of the aforementioned factors. Additionally, BCL2L12 mRNA expression status was also found to be an independent predictor of favorable prognosis, when multivariate analysis was adjusted for either ER (hazard ratio [HR] = 0.15; 95% confidence interval [95% CI] = 0.03–0.72; p = 0.018) or PR (hazard ratio [HR] = 0.17; 95% confidence interval [95% CI] = 0.03–0.85; p = 0.031). Interestingly, BCL2L12 expression status remained
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
4
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
Table 2 Associations between BCL2L12 status and clinicopathological variables. No. of patients (%) Variable
Total
BCL2L12-Lowa
BCL2L12-Higha
p value
Tumor size ≤2 cm 2–5 cm N5 cm x
20 64 13 21
8 (40.0) 35 (54.7) 11 (84.6)
12 (60.0) 29 (45.3) 2 (15.4)
0.04d
Metastasis Negative Positive x
86 16 16
46 (53.5) 14 (87.5)
40 (46.5) 2 (12.5)
0.012e
Nottingham prognostic index Poor 42 Moderate 40 Good 10 x 26
28 (66.7) 18 (45.0) 5 (50.0)
14 (33.3) 22 (55.0) 5 (50.0)
0.13d
Stageb I/II III/IV x
69 30 19
33 (47.8) 23 (76.7)
36 (52.2) 7 (23.3)
0.009e
Age ≤35 35–55 N55 x
3 36 66 13
2 (66.7) 14 (38.9) 44 (66.7)
1 (33.3) 22 (61.1) 22 (33.3)
0.024d
35 (58.3) 21 (51.2)
25 (41.7) 20 (48.8)
0.54e
Histological gradec I/II 60 III 41 x 17
x: Unknown status. Statistically significant results are in bold type. a Cutoff point: 0.77 RQ units equals to 55th percentile. b TNM staging system. c Bloom–Scarff–Richardson grading system. d Pearson chi-square test. e Fisher's exact test.
a significant predictor of improved DFS, in three reduced models consisting of several of the aforementioned molecular and clinicopathological parameters. In more detail, BCL2L12 was found to be an independent prognostic marker according to the multivariate Cox regression models, which included either patients' age, NPI and ER status (hazard ratio [HR] = 0.10, 95% confidence interval [95% CI] = 0.018–0.56, p = 0.009), or patients' age, NPI and PR status (hazard ratio [HR] = 0.14; 95% confidence interval [95% CI] = 0.024–0.81; p = 0.029), or tumor size, histological grade and patients' age (hazard ratio [HR] = 0.12, 95% confidence interval [95% CI] = 0.015–0.97; p = 0.047).
Fig. 3. Kaplan–Meier curves for (A) disease-free survival (DFS) and (B) overall survival (OS) of BC patients with tumors demonstrating high or low BCL2L12 expression.
Discussion It is well known that members of the BCL2 family are implicated in cancer pathogenesis and progression. Therefore, plenty of these apoptosis-related genes and proteins have been studied in many malignancies, including BC, for their prognostic and predictive potential [26]. For example, ERa-negative BC patients exhibit decreased expression of BAK, a member of the BCL2 member, suggesting that BAK down-
Fig. 2. Box plots representing BCL2L12 expression levels in relation to (A) disease stage and (B) to tumor size. The p value was calculated by the “Mann–Whitney U test” and “Jonckheere– Terpstra test”, respectively. The box refers to the range defined by the 25th and the 75th percentile and the bold line corresponds to the median value.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx Table 3 Cox univariate regression analysis of BCL2L12 regarding disease-free survival. Univariate analysis for disease-free survival ( N = 101) Variable BCL2L12 Low expression High expression Grade I/II III Size (Ordinal) Age (Ordinal) NPI (Ordinal) ER Status Negative Positive PR status Negative Positive HER-2 status (Ordinal)
HR
95% CI
p valuea
1.00 0.19
0.04–0.85
0.030
1.00 0.99 1.59 2.62 0.76
0.41–2.44 0.71–3.60 0.92–7.49 0.37–1.57
0.99 0.26 0.07 0.46
1.00 0.17
0.05–0.54
0.003
1.00 0.02 1.18
0.00–1.19 0.81–1.72
0.06 0.39
Abbreviations: HR = hazard ratio; CI = confidence intervals; NPI = Nottingham prognostic index. Statistically significant results are in bold type. a Test for trend.
regulation is associated with malignant development in the mammary gland [27]. On the other hand, BCL2, the prototypic member of the BCL2 family, appears to be a powerful independent favorable prognostic factor that can significantly influence the current prognostic models [20, 28–31], leading to its incorporation into the 21-gene panel of Oncotype DX [6]. The favorable prognostic nature of BCL2 in BC, is reinforced by its positive correlation with ER and PR expression [28,32,33], whereas loss
Table 4 Cox multivariate regression analysis of BCL2L12 regarding disease-free survival. Multivariate analysis for disease-free survival ( N = 89) Full modela HR
95% CI
p valueg
1.00 0.12
0.015–0.98
0.048
1.00 1.05 1.03 1.62 0.67
0.32–3.47 0.35 – 3.02 0.49–5.37 0.26–1.71
0.94 0.96 0.43 0.40
Variable
HR
95% CI
p value⁎
BCL2L12 Model 1b Model 2c Model 3d Model 4e Model 5f Model 6g
0.15 0.17 0.23 0.12 0.10 0.14
0.03–0.72 0.03–0.85 0.05–1.03 0.015–0.97 0.018–0.56 0.024–0.81
0.018 0.031 0.054 0.047 0.009 0.029
Variable BCL2L12 Low expression High expression Grade I/II III Size (Ordinal) Age (Ordinal) Nottingham Prognostic Index (Ordinal) Reduced models
Abbreviations: HR = hazard ratio; CI = confidence intervals; NPI = Nottingham prognostic index. Statistically significant results are in bold type. a Multivariate model adjusted for patients' age, tumor size, histological grade, and Nottingham prognostic index. b Multivariate model adjusted for ER status. c Multivariate model adjusted for PR status. d Multivariate model adjusted for HER-2 status. e Multivariate model adjusted for patients' age, tumor size and histological grade. f Multivariate model adjusted for patients' age, NPI and ER status. g Multivariate model adjusted for patients' age, NPI and PR status. ⁎ Test for trend.
5
of BCL2 expression is associated with the presence of unfavorable prognostic molecular markers (e.g. EGFR, HER2, and p53) [33]. BCL2L12 is a novel member of the BCL2 family, which is subjected to alternative splicing resulting in the generation of 13 spliced variants. Recently, the expression profile of eight novel BCL2L12 transcript variants was examined in several cancer cell lines and significant variations in their expression profile were observed. Regarding BC, BCL2L12 variants 4, 5, 10, 11 and 13 demonstrated differential expression among the MCF-7, BT474 and BT-20 cell lines [34]. The BCL2L12 mRNA expression has also been examined in tissue samples obtained from various malignancies, suggesting its possible prognostic role in cancer. As far as BC is concerned, patients bearing BCL2L12-positive tumors seem to have significantly longer DFS and OS [19,20]. In view of the above mentioned data we investigated the possible prognostic role of BCL2L12 in a large cohort of cancerous and non-cancerous breast specimens, using a highly sensitive qRT-PCR method. Based on our results, increased BCL2L12 expression levels were significantly associated with early TNM stages (p = 0.009), absence of metastasis (p = 0.012) and small tumor size (p = 0.04), supporting the previously reported hypothesis that mRNA expression of BCL2L12 is associated with less aggressive forms of BC [19,20]. Tumor size is an independent prognostic factor for BC and its increase is accompanied by decreased survival rates and high risk of recurrence [35–37]. The correlation of BCL2L12 expression with tumor size is interesting and may prove useful for the identification of women with favorable prognosis. Moreover, elevated levels of BCL2L12 mRNA were observed in patients 35–55 years old (p = 0.024). Epidemiological studies suggest an exponential increase of this malignancy until the age of 50 followed by a slight decrease around menopause age and then a linear rise, a phenomenon known as Clemmensen hook [38–40]. Many prospective and retrospective studies suggest that younger (≤35 years old) BC patients exhibit worse clinical outcome compared to the older ones (40–69 years old), having a significant higher recurrence risk [38,41–43]. Furthermore, although limited, there are data implying that age is an independent marker of poor prognosis in BC [43]. According to our results, women belonging to the age group characterized by low incidence rates and good prognosis, namely 35–55 years old, exhibited increased levels of BCL2L12 expression, enhancing the favorable nature of this gene. According to Kaplan–Meier (p = 0.015) and univariate Cox regression analysis (p = 0.030), there is a significant association of BCL2L12 overexpression with superior DFS intervals, supporting the favorable prognostic role of this gene. In more detail, patients with high BCL2L12 expression were almost five times less likely to relapse in relation to women with decreased BCL2L12 mRNA levels. Furthermore, multivariate regression analysis, adjusted for non-molecular parameters, namely, histological grade, tumor size, patients' age and Nottingham prognostic index, revealed that BCL2L12 expression can independently predict good recurrence-free survival (p = 0.048). In addition, a reduced model consisting of the same parameters, but excluding the Nottingham prognostic index, highlighted the significance of BCL2L12 mRNA expression as an independent predictor of DFS (p = 0.047). Since measurement of ER, PR and HER-2 is an established clinical practice for BC patients' prognosis, we deemed the incorporation of these markers in multiparametric logistic regression models necessary. A multivariate model including BCL2L12 expression profile and ER status, revealed that increased levels of BCL2L12 can independently predict a superior DFS (hazard ratio [HR] = 0.15; 95% confidence interval [95% CI] = 0.03–0.72; p = 0.018). Similar results were observed when multiparametric models were adjusted for PR status (hazard ratio [HR] = 0.17; 95% confidence interval [95% CI] = 0.03–0.85; p = 0.031) but not for HER-2 expression profile (hazard ratio [HR] = 0.23; 95% confidence interval [95% CI] = 0.05–1.03; p = 0.054). Finally, two additional reduced models adjusted for patients' age, NPI and ER (p = 0.009) or PR (p = 0.029) status revealed that BCL2L12 mRNA expression is a strong independent prognostic marker for BC.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
6
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
Conclusion A possible prognostic role of BCL2L12 in BC has been revealed by previous experimental data generated from conventional PCR. In this study BCL2L12 expression was quantified in a large cohort of BC patients with a long-term follow up, using a sensitive qRT-PCR approach. According to our results, BCL2L12 is an independent favorable prognostic marker in BC, suggesting its use in the assessment of relapse risk in this type of cancer. However, the current study represents a first attempt at assessing the role of BCL2L12 as a potential tumor marker in BC and future studies, in an independent cohort, are needed for validation of our results and to further demonstrate its role in clinical practice. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2014.09.008. Conflict of Interest We declare that we have no conflict of interest Acknowledgements This research has been co-financed by the European Union (European Social Fund; ESF) and Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF), Research Funding Program: THALES, Investing in Knowledge Society through the European Social Fund (THALES UoA-Biopromo, MIS 377046). References [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69–90. [2] Polyak K. Heterogeneity in breast cancer. J Clin Invest 2011;121:3786–8. [3] Kittaneh M, Montero AJ, Gluck S. Molecular profiling for breast cancer: a comprehensive review. Biomark Cancer 2013;5:61–70. [4] Ellsworth RE, Decewicz DJ, Shriver CD, Ellsworth DL. Breast cancer in the personal genomics era. Curr Genomics May 2010;11(3):146–61. http://dx.doi.org/10.2174/ 138920210791110951. [5] Tian S, Roepman P, Van't Veer LJ, Bernards R, de Snoo F, Glas AM. Biological functions of the genes in the mammaprint breast cancer profile reflect the hallmarks of cancer. Biomark Insights 2010;5:129–38. [6] Cronin M, Sangli C, Liu ML, Pho M, Dutta D, Nguyen A, et al. Analytical validation of the oncotype dx genomic diagnostic test for recurrence prognosis and therapeutic response prediction in node-negative, estrogen receptor-positive breast cancer. Clin Chem 2007;53:1084–91. [7] Parker JS, Mullins M, Cheang MC, Leung S, Voduc D, Vickery T, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol Mar. 10 2009; 27(8):1160–7. http://dx.doi.org/10.1200/JCO.2008.18.1370 [Epub 2009 Feb 9. 2009]. [8] Scorilas A, Kyriakopoulou L, Yousef GM, Ashworth LK, Kwamie A, Diamandis EP. Molecular cloning, physical mapping, and expression analysis of a novel gene, bcl2l12, encoding a proline-rich protein with a highly conserved bh2 domain of the bcl-2 family. Genomics 2001;72:217–21. [9] Yousef GM, Kopolovic AD, Elliott MB, Diamandis EP. Genomic overview of serine proteases. Biochem Biophys Res Commun 2003;305(1):28–36. [10] Scorilas A, Chiang PM, Katsaros D, Yousef GM, Diamandis EP. Molecular characterization of a new gene, ceal1, encoding for a carcinoembryonic antigen-like protein with a highly conserved domain of eukaryotic translation initiation factors. Gene May 22 2003;310:79–89 [2003]. [11] Borglum AD, Byskov A, Ragno P, Roldan AL, Tripputi P, Cassani G, et al. Assignment of the urokinase-type plasminogen activator receptor gene (plaur) to chromosome 19q13.1-q13.2. Am J Hum Genet Mar. 1992;50(3):492–7 [1992]. [12] Florou D, Papadopoulos IN, Scorilas A. Molecular analysis and prognostic impact of the novel apoptotic gene bcl2l12 in gastric cancer. Biochem Biophys Res Commun 2010;391:214–8. [13] Foutadakis S, Avgeris M, Tokas T, Stravodimos K, Scorilas A. Increased bcl2l12 expression predicts the short-term relapse of patients with tat1 bladder cancer following transurethral resection of bladder tumors. Urol Oncol 2014;32:39.e29–36. [14] Fendri A, Kontos CK, Khabir A, Mokdad-Gargouri R, Scorilas A. Bcl2l12 is a novel biomarker for the prediction of short-term relapse in nasopharyngeal carcinoma. Mol Med 2011;17:163–71. [15] Geomela PA, Kontos CK, Yiotakis I, Scorilas A. Quantitative expression analysis of the apoptosis-related gene, bcl2l12, in head and neck squamous cell carcinoma. J Oral Pathol Med 2013;42:154–61.
[16] Kontos CK, Papadopoulos IN, Scorilas A. Quantitative expression analysis and prognostic significance of the novel apoptosis-related gene bcl2l12 in colon cancer. Biol Chem 2008;389:1467–75. [17] Thomadaki H, Floros KV, Pavlovic S, Tosic N, Gourgiotis D, Colovic M, et al. Overexpression of the novel member of the bcl2 gene family, bcl2l12, is associated with the disease outcome in patients with acute myeloid leukemia. Clin Biochem 2012; 45:1362–7. [18] Papageorgiou SG, Kontos CK, Pappa V, Thomadaki H, Kontsioti F, Dervenoulas J, et al. The novel member of the bcl2 gene family, bcl2l12, is substantially elevated in chronic lymphocytic leukemia patients, supporting its value as a significant biomarker. Oncologist 2011;16:1280–91. [19] Talieri M, Diamandis EP, Katsaros N, Gourgiotis D, Scorilas A. Expression of bcl2l12, a new member of apoptosis-related genes, in breast tumors. Thromb Haemost 2003; 89:1081–8. [20] Thomadaki H, Talieri M, Scorilas A. Prognostic value of the apoptosis related genes bcl2 and bcl2l12 in breast cancer. Cancer Lett 2007;247:48–55. [21] McShane LM, Altman DG, Sauerbrei W, Taube SE, Gion M, Clark GM. Reporting recommendations for tumor marker prognostic studies (remark). Breast Cancer Res Treat 2006;100:229–35. [22] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative pcr and the 2(−delta delta c(t)) method. Methods Dec. 2001;25(4): 402–8 [2001]. [23] de Kok JB, Roelofs RW, Giesendorf BA, Pennings JL, Waas ET, Feuth T, et al. Normalization of gene expression measurements in tumor tissues: comparison of 13 endogenous control genes. Lab Invest Jan. 2005;85(1):154–9 [2005]. [24] Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology Apr. 1982;143(1):29–36 [1982]. [25] Camp RL, Dolled-Filhart M, Rimm DL. X-tile: a new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clin Cancer Res 2004;10: 7252–9. [26] Thomadaki H, Scorilas A. Bcl2 family of apoptosis-related genes: functions and clinical implications in cancer. Crit Rev Clin Lab Sci 2006;43:1–67. [27] Eguchi H, Suga K, Saji H, Toi M, Nakachi K, Hayashi SI. Different expression patterns of bcl-2 family genes in breast cancer by estrogen receptor status with special reference to pro-apoptotic bak gene. Cell Death Differ 2000;7:439–46. [28] Hwang KT, Woo JW, Shin HC, Kim HS, Ahn SK, Moon HG, et al. Prognostic influence of bcl2 expression in breast cancer. Int J Cancer 2012;131:E1109–19. [29] Dawson SJ, Makretsov N, Blows FM, Driver KE, Provenzano E, Le Quesne J, et al. Bcl2 in breast cancer: a favourable prognostic marker across molecular subtypes and independent of adjuvant therapy received. Br J Cancer 2010;103:668–75. [30] Parton M, Dowsett M, Smith I. Studies of apoptosis in breast cancer. BMJ 2001;322: 1528–32. [31] Sjostrom J, Blomqvist C, von Boguslawski K, Bengtsson NO, Mjaaland I, Malmstrom P, et al. The predictive value of bcl-2, bax, bcl-xl, bag-1, fas, and fasl for chemotherapy response in advanced breast cancer. Clin Cancer Res 2002;8:811–6. [32] Kolacinska A, Chalubinska J, Zawlik I, Szymanska B, Borowska-Garganisz E, Nowik M, et al. Apoptosis-, proliferation, immune function-, and drug resistance- related genes in ER positive, her2 positive and triple negative breast cancer. Neoplasma 2012;59: 424–32. [33] Hellemans P, van Dam PA, Weyler J, van Oosterom AT, Buytaert P, Van Marck E. Prognostic value of bcl-2 expression in invasive breast cancer. Br J Cancer 1995; 72:354–60. [34] Kontos CK, Scorilas A. Molecular cloning of novel alternatively spliced variants of bcl2l12, a new member of the bcl2 gene family, and their expression analysis in cancer cells. Gene Aug. 15 2012;505(1):153–66. http://dx.doi.org/10.1016/j.gene.2012. 04.084 [Epub 2012 Jun 1. 2012]. [35] Carter CL, Allen C, Henson DE. Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases. Cancer 1989;63:181–7. [36] Cianfrocca M, Goldstein LJ. Prognostic and predictive factors in early-stage breast cancer. Oncologist 2004;9:606–16. [37] Narod SA. Tumour size predicts long-term survival among women with lymph node-positive breast cancer. Curr Oncol 2012;19:249–53. [38] Nixon AJ, Neuberg D, Hayes DF, Gelman R, Connolly JL, Schnitt S, et al. Relationship of patient age to pathologic features of the tumor and prognosis for patients with stage I or II breast cancer. J Clin Oncol 1994;12:888–94. [39] Xiong Q, Valero V, Kau V, Kau SW, Taylor S, Smith TL, et al. Female patients with breast carcinoma age 30 years and younger have a poor prognosis: The M.D. Anderson Cancer Center experience. Cancer 2001;92:2523–8. [40] Kroman N, Jensen MB, Wohlfahrt J, Mouridsen HT, Andersen PK, Melbye M. Factors influencing the effect of age on prognosis in breast cancer: population based study. BMJ 2000;320:474–8. [41] Keating Litton J. Is age still the deciding factor? Commentary on “The effect of age on delay in diagnosis and stage of breast cancer” by Partridge et al. Oncologist 2012;17: 744–6. [42] Assi HA, Khoury KE, Dbouk H, Khalil LE, Mouhieddine TH, El Saghir NS. Epidemiology and prognosis of breast cancer in young women. J Thorac Dis 2013;5:S2–8. [43] Copson E, Eccles B, Maishman T, Gerty S, Stanton L, Cutress RI, et al. Prospective observational study of breast cancer treatment outcomes for UK women aged 18–40 years at diagnosis: the POSH Study. J Natl Cancer Inst 2013;105:978–88.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
BCL2L12: A promising molecular prognostic biomarker in breast cancer Alexandros Tzovaras a, Athina Kladi-Skandali b, Kleita Michaelidou b, George C. Zografos c, Ioannis Missitzis d, Alexandros Ardavanis a, Andreas Scorilas b,⁎ a
First Department of Medical Oncology, “Saint Savvas”, Anticancer Hospital, 11522, Athens, Greece Department of Biochemistry and Molecular Biology, University of Athens, 15701 Athens, Greece First Propaedeutic Surgical Department, Athens Medical School, “Hippokration” General Hospital, Athens, Greece d The Breast Clinic “Saint Savvas”, Anticancer Hospital, 11522,Athens, Greece b c
a r t i c l e
i n f o
Article history: Received 1 July 2014 Received in revised form 3 September 2014 Accepted 4 September 2014 Available online xxxx Keywords: BCL2 family Apoptosis Molecular tumor markers Prognostic biomarker Quantitative real-time PCR
a b s t r a c t Objectives: BCL2-like 12 (BCL2L12) is a new member of the BCL2 gene family that was discovered and cloned by members of our group and found to be expressed in the mammary gland. Many genes of the BCL2 family were found to be implicated in breast carcinogenesis and to serve as possible prognostic markers. The aim of the present study was the quantification of BCL2L12 mRNA expression in order to assess its value as a prognostic tissue biomarker in breast cancer (BC). Design and methods: BCL2L12 mRNA levels were determined in a statistically significant sample size of cancerous (N = 108) and adjacent non-cancerous (N = 71) breast tissues using a highly sensitive quantitative real-time polymerase chain reaction (qRT-PCR) method. Relative quantification analysis was conducted using the comparative CT (2− ΔΔCT) method, whereas the association between BCL2L12 expression and clinopathological data, disease-free survival (DFS) and overall survival (OS) were estimated by statistical analysis. Results: BCL2L12 mRNA expression was decreased in malignant samples compared to the histologically normal counterparts (p = 0.012). Significant relationships between BCL2L12 expression and TNM stages (p = 0.009), metastatic potential (p = 0.012), tumor size (p = 0.04) and age (p = 0.024) were observed. Moreover, Kaplan–Meier and Cox univariate analyses indicated that BCL2L12 expression is associated with longer DFS, whereas multivariate analysis pointed out the independent favorable prognostic value of BCL2L12. Conclusions: According to our results, BCL2L12 mRNA expression is a favorable prognostic marker of DFS for BC patients, suggesting its possible application as a novel prognostic indicator of this malignancy. © 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Breast cancer (BC), the leading cause of cancer death in women worldwide [1], is a multifactorial and particularly heterogeneous malignancy [2]. In the genomic era, the traditional clinical–pathological approach of BC prognosis seems inadequate to describe the disease complexity and therefore molecular biomarker tests as well as gene expression signatures constantly gain ground in BC patients' management [3,4]. Nowadays, two real-time PCR-based assays (Oncotype DX and PAM50) and a microarray-based multigene test (MammaPrint) are commercially available and enable the accurate classification of BC patients, estimation of recurrence risk and prediction of treatment
Abbreviations: BC, breast cancer; qRT-PCR, quantitative real-time polymerase chain reaction;DFS,disease-free survival; OS, overallsurvival; RQ units,relative quantification units; CT, threshold cycle; NPI, Nottingham Prognostic Index. ⁎ Corresponding author at: Department of Biochemistry and Molecular Biology, University of Athens, 15701 Panepistimiopolis, Athens, Greece. Fax: +30 210 727 4158. E-mail address: [email protected] (A. Scorilas).
response [3]. These approaches allow the simultaneous analysis of the expression patterns of a set of genes, associated with several hallmarks of BC including angiogenesis, metastasis and impairment of apoptosis [5–7]. BCL2-like 12 (BCL2L12) is a novel member of the BCL2 family of apoptosis-related genes that maps to chromosome 19q13.3 [8], a genomic locus where many cancer-related genes are located, such as the KLK gene family [9], the CEACAM/PSG cluster [10] and the gene coding for the urokinase-type plasminogen activator receptor (uPAR) [11]. The BCL2L12 mRNA expression has been studied not only in solid tumors [12–16] but also in haematological malignancies [17,18], and the results are encouraging regarding the potential use of this gene as a tumor marker. As far as BC is concerned, preliminary data generated from conventional PCR supports the notion that this apoptosis-related gene is a favorable prognostic marker respecting BC patients' DFS and OS [19,20]. In light of these data, we conducted this study aiming at the verification of the prognostic value of this gene in BC. For this purpose, a sensitive quantitative RT-PCR (qRT-PCR) methodology was developed and applied for the expression analysis of BCL2L12 in a large number of BC
http://dx.doi.org/10.1016/j.clinbiochem.2014.09.008 0009-9120/© 2014 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
2
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
patients. To the best of our knowledge, this is the first study examining the expression of BCL2L12 gene in breast carcinoma tissues using this methodology. Materials and methods Breast tissue specimens The current study was designed according to new guidelines for reporting new tumor biomarkers [21] and was approved by the institutional review boards of the Anticancer Hospital “Saint Savvas”. Research procedures of this study comply with the ethical standards of the World Medical Association Declaration of Helsinki (version: 2008). Informed consent was obtained from patients participating in the study. The study group consisted of 179 samples (71 pairs of BC tissues and their normal counterparts and 37 additional samples of tumoral tissue), obtained from women who had undergone surgery for histologically verified primary breast carcinomas. Patients were nursed at the 1st Oncology Department of the Anticancer Hospital “Saint Savvas”, Athens, Greece and sample collection took place between 2009 and 2011. All tissue specimens were snap-frozen in liquid nitrogen and preserved at − 80 °C until use. Adequate amounts of tissue for RNA extraction and detailed clinicopathological data (Supplementary Table S1) were prerequisites for specimens' further analysis. Patient age ranged from 31 up to 87 years (mean ± SD, 59.20 ± 12.83) and tumor size varied between 0.8 and 10.5 cm (mean ± SD, 3.20 ± 1.82). The median follow-up period for DFS and OS was 32.12 and 37.51 months, respectively, with a range of 0.60–47.44 months (Supplementary Table S2). RNA extraction and first-strand cDNA synthesis Total RNA was extracted from frozen pulverized breast tissue as well as from MCF-7 cells (human breast adenocarcinoma cell line) using the guanidine isothiocyanate method (TRI Reagent®, Ambion). According to the standard protocol, the RNA pellet was diluted in RNA storage solution (Ambion) and stored immediately at −80 °C until further analysis. The concentration and purity of the RNA were determined by UVspectrophotometry, while RNA quality was verified by agarose gel electrophoresis. Thereafter, first-strand cDNA was synthesized from 2 μg of total RNA according to the previously described protocol [13]. Quantitative real-time PCR (qRT-PCR) A sensitive qRT-PCR methodology (Supplementary Fig. 1) was developed for the molecular analysis of BCL2L12, utilising the SYBR Green I chemistry. Based on NCBI information, two pairs of genespecific primers were designed for the amplification of the endogenous reference gene (HPRT1; GenBank accession number: NM_000194.2) and the target gene (BCL2L12; GenBank accession number: NM_138639.1). The sequences of the BCL2L12 qRT-PCR primers were: 5′-CCCTCGGCCTTGCTCTCT-3′ and 5′-TCCGCAGTATGGCTTCCTTC-3′, producing a 182-bp PCR amplicon. The corresponding sequences of the HPRT1 primers were 5'-TGGAAAGGGTGTTTATTCCTCAT-3' and 5'ATGTAATCCAGCAGGTCAGCAA-3' giving a 151 bp PCR product. qRT-PCR for HPRT1 and BCL2L12 cDNAs was conducted in duplicate, on the same 96-well PCR microplate in a 7500 Thermal Cycler (Applied Biosystems). The 10 μl reaction mixture contained 10 ng of cDNA, 5 μl Kapa SYBR Fast Universal 2× qPCR Master Mix (Kapa Biosystems), 0.2 μl of 50 × Rox Low passive reference dye (Kapa Biosystems), 50 nM of each BCL2L12 primer/75 nM of each HPRT1 primer and DEPC-treated water to the final volume. The PCR protocol conditions consisted of an initial 3-minute polymerase activation step at 95 °C, followed by 40 cycles of 95 °C for 15 seconds for template denaturation, and 60 °C for 1 minute for primer annealing and extension. After completion of the PCR reaction, a melt curve analysis was conducted to
distinguish the PCR products of interest from any non-specific products and/or primer–dimers through their particular melting temperatures (Tm). The Tm of the BCL2L12 and HPRT1 PCR products were 84.5 °C and 81.0 °C, respectively. The relative quantification of the expression analysis of all breast specimens was conducted using the comparative CT (2−ΔΔCT) method, and the expression levels of the gene of interest are expressed as relative quantification units (RQ units). According to this approach, ΔCT represents the difference between the threshold cycle (CT) of the target gene (BCL2L12) and the CT value of the corresponding endogenous reference gene (HPRT1) in each sample under study as follows: (ΔCT = CT BCL2L12 − CT HPRT1). The ΔΔCT value is equal to the difference between the average ΔCT value of an experimental sample and the average ΔCT of the corresponding calibrator according to the formula (ΔCT test sample − ΔCT calibrator). The application of the comparative CT method requires approximately equal amplification efficiencies for target and reference genes [22]. For this reason, data from previously described validation experiments were used [13]. In the current study, HPRT1 was used as an internal control gene for normalization purposes, since it is one of the most stable housekeeping genes and its use as a single reference gene is recommended for expression studies in several types of cancer, including BC [23]. Moreover, the human BC cell line MCF-7 served as calibrator. Statistical analyses The nonparametric Wilcoxon signed-rank test was performed in order to estimate the differences in BCL2L12 expression levels between cancerous and non-cancerous tissues. To evaluate the discriminatory value of BCL2L12 expression levels between tumoral and normal tissues, receiver operating characteristic curve (ROC) was constructed and the area under the ROC curve (AUC) was analyzed by the Hanley and McNeil method [24]. In order to examine the relation of BCL2L12 expression levels, as a continuous variable, with various qualitative clinicopathological parameters, we performed either Mann–Whitney U test or Jonckheere–Terpstra test. Since there are no established cutoffs for BCL2L12 expression, the X-Tile algorithm was used to produce an optimal bifurcating point of 0.77 RQ Units (equal to 55th percentile) [25]. According to the results, the patient cohort was classified into high and low BCL2L12 expression groups and associations between this status and other categorical clinicopathological variables were analyzed using either Fisher's exact test or Pearson chi-square test. The associations between various markers of prognosis and the relative risks for disease recurrence and death of patients were determined using univariate and multivariate Cox proportional hazard regression models. The resulting p value was evaluated by the test for trend approach. The survival analysis included also the construction of Kaplan–Meier curves, which represent the percentage probability of the DFS and OS, using as threshold the aforementioned cutoff. The p value was calculated using the long-rank test. Finally, the level of significance was defined at a probability value of less than 0.05 (p b 0.05). Results Evaluation of the discriminatory potential of BCL2L12 mRNA expression in BC patients According to the descriptive statistical analysis, BCL2L12 expression was remarkably elevated in non-cancerous tissues compared to their malignant counterparts. In particular, BCL2L12 mRNA expression in the BC samples ranged from 0.02 up to 28.25 RQ units with a mean ± SD of 1.63 ± 3.83 RQ units and a median of 0.60 RQ units. On the other hand, BCL2L12 mRNA expression in the adjacent histologically normal tissues varied between 0.02 and 70.00 RQ units with a mean ± SD of
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
5.00 ± 10.00 RQ units and a median of 0.91 RQ units (Supplementary Table 2). The comparison of BCL2L12 mRNA levels in 71 pairs of breast tissues (normal versus cancerous tissue from the same patient) revealed a significantly higher expression in nonmalignant samples than in their matched tumoral tissues (p = 0.012). According to Fig. 1, 29 patients (40.8%) demonstrated increased expression levels in the malignant breast tissue compared to the normal counterpart, whereas in 39 women (55.0%), BCL2L12 expression in the cancerous specimen was lower than in the paired non-cancerous tissue. In order to assess the diagnostic value of BCL2L12 expression, we performed receiver operating characteristic (ROC) curve analysis (Supplementary Fig. S2). BCL212 could distinguish patients with BC from healthy controls, exhibiting, however, a moderate differential diagnostic performance with an AUC of 0.65 (AUC = 0.65; 95% CI = 0.56–0.74; p = 0.001). Correlations between BCL2L12 mRNA expression and BC patients' clinicopathological parameters Associations between BCL2L12 expression and patients' clinicopathological parameters such as tumor size and histological grade, molecular subtype, stage and metastasis were examined (Tables 1 and 2), revealing that BCL2L12 overexpression was found to be associated with favorable clinicopathological features. Women with early TNM stage (I/II) BC demonstrated significantly (p = 0.007) higher BCL2L12 mRNA levels (median = 0.84 RQ units) in comparison to those with stage III/IV BC (median = 0.41 RQ units; Fig. 2A). Moreover, an inverse relationship between BCL2L12 expression levels and tumor size was observed (p = 0.010). Particularly, median BCL2L12 expression in the group of patients with tumor size ≤2 cm was 1.01 RQ units, whereas in tumors of 2–5 cm and N5 cm, the median values were 0.75 and 0.22 RQ units, respectively (Fig. 2B). Finally, women in the age range of 35 to 55 years exhibited slightly higher, although not significant (p = 0.07), BCL2L12 expression (median = 1.02 RQ units) than patients aged less than 35 years (median = 0.46 RQ units) and those aged over 55 years (median = 0.48 RQ units; Table 1). Similar results were obtained when we classified patients into two categories (BCL2L12-high and BCL2L12-low) on the basis of BCL2L12 mRNA levels, and we examined associations between this status and the abovementioned categorical variables (Table 2). For this reason, a cutoff value of 0.77 RQ units (equal to the 55th percentile) was generated and the associations were examined using the Pearson chi-square and Fisher exact tests, wherever appropriate. According to these results, BCL2L12 overexpression was significantly associated with early TNM
Fig. 1. Comparison of BCL2L12 mRNA levels in malignant and histologically normal counterparts. Bar graph representing the difference in the BCL2L12 expression between cancerous and non-cancerous breast specimens. The p value was calculated using “Wilcoxon signed-rank test”.
3
Table 1 BCL2L12 mRNA expression in BC patients according to classic clinicopathological data. Variable
No. of patients
Mean ± SD
Median
p value
Histological gradea I II III
3 57 41
3.43 ± 5.40 1.74 ± 4.30 1.56 ± 3.41
0.42 0.58 0.77
0.17c
Stageb I/II III/IV
69 30
2.18 ± 4.68 0.72 ± 0.88
0.84 0.41
0.007d
Tumor size ≤2 cm 2–5 cm N5 cm
20 64 13
2.39 ± 5.10 1.84 ± 4.05 0.41 ± 0.41
1.01 0.75 0.22
0.010c
Molecular subtype Luminal A Luminal B Triple negative HER-2 group
32 27 16 8
2.15 1.56 1.03 1.84
5.43 4.18 0.72 2.55
0.59 0.52 0.85 0.65
0.24c
Type NST Lobular
74 13
1.20 ± 2.56 1.97 ± 4.47
0.42 0.75
0.20d
Age ≤35 35–55 N55
3 36 66
0.50 ± 0.40 2.47 ± 5.12 1.28 ± 3.07
0.46 1.02 0.48
0.07c
± ± ± ±
Abbreviations: SD = standard deviation. Statistically significant results are shown in bold type. a Bloom–Scarff–Richardson grading system. b TNM staging system. c Calculated by the “Jonckheere–Terpstra test”. d Calculated by the “Mann–Whitney U test”.
stages (p = 0.009), small tumor size (p = 0.04) and age range of 35–55 years (p = 0.024). Regarding BCL2L12 expression and the presence of metastases, a statistically significant negative association was demonstrated (p = 0.012). BCL2L12 mRNA status and survival of BC patients Complete follow-up data were available for 101 patients, among whom 16 (15.8%) relapsed and 7 (6.9%) died. The prognostic significance of BCL2L12 expression in our cohort of BC patients was estimated by the Kaplan–Meier survival analysis (Fig. 3) and the Cox regression (Tables 3 and 4), after classification of patients into two categories, based on BCL2L12 expression as described in the “Statistical analysis” section. According to Kaplan–Meir analysis, patients overexpressing BCL2L12 have significantly superior DFS (p = 0.015) and OS (p = 0.025; median DFS and OS time = 37.5 and 38.2 months, respectively) compared to those with decreased BCL2L12 mRNA levels (median DFS and OS time = 30.6 and 34.7 months, respectively). Cox univariate regression analysis (Table 3), also, revealed the favorable prognostic value of BCL2L12 overexpression in terms of DFS (p = 0.030), but not of OS (data not shown), as women demonstrating increased BCL2L12 expression had a significantly decreased risk of relapse (hazard ratio [HR] = 0.19, 95% confidence interval [95% CI] = 0.44–0.85; p = 0.030). In the Cox multivariate analysis (Table 4), the full multiparametric model was adjusted for tumor size, histological grade, patients' age and Nottingham prognostic index and revealed that BCL2L12 mRNA expression (hazard ratio [HR] = 0.12; 95% confidence interval [95% CI] = 0.015–0.98; p = 0.048) can predict BC patients' relapse independently of the aforementioned factors. Additionally, BCL2L12 mRNA expression status was also found to be an independent predictor of favorable prognosis, when multivariate analysis was adjusted for either ER (hazard ratio [HR] = 0.15; 95% confidence interval [95% CI] = 0.03–0.72; p = 0.018) or PR (hazard ratio [HR] = 0.17; 95% confidence interval [95% CI] = 0.03–0.85; p = 0.031). Interestingly, BCL2L12 expression status remained
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
4
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
Table 2 Associations between BCL2L12 status and clinicopathological variables. No. of patients (%) Variable
Total
BCL2L12-Lowa
BCL2L12-Higha
p value
Tumor size ≤2 cm 2–5 cm N5 cm x
20 64 13 21
8 (40.0) 35 (54.7) 11 (84.6)
12 (60.0) 29 (45.3) 2 (15.4)
0.04d
Metastasis Negative Positive x
86 16 16
46 (53.5) 14 (87.5)
40 (46.5) 2 (12.5)
0.012e
Nottingham prognostic index Poor 42 Moderate 40 Good 10 x 26
28 (66.7) 18 (45.0) 5 (50.0)
14 (33.3) 22 (55.0) 5 (50.0)
0.13d
Stageb I/II III/IV x
69 30 19
33 (47.8) 23 (76.7)
36 (52.2) 7 (23.3)
0.009e
Age ≤35 35–55 N55 x
3 36 66 13
2 (66.7) 14 (38.9) 44 (66.7)
1 (33.3) 22 (61.1) 22 (33.3)
0.024d
35 (58.3) 21 (51.2)
25 (41.7) 20 (48.8)
0.54e
Histological gradec I/II 60 III 41 x 17
x: Unknown status. Statistically significant results are in bold type. a Cutoff point: 0.77 RQ units equals to 55th percentile. b TNM staging system. c Bloom–Scarff–Richardson grading system. d Pearson chi-square test. e Fisher's exact test.
a significant predictor of improved DFS, in three reduced models consisting of several of the aforementioned molecular and clinicopathological parameters. In more detail, BCL2L12 was found to be an independent prognostic marker according to the multivariate Cox regression models, which included either patients' age, NPI and ER status (hazard ratio [HR] = 0.10, 95% confidence interval [95% CI] = 0.018–0.56, p = 0.009), or patients' age, NPI and PR status (hazard ratio [HR] = 0.14; 95% confidence interval [95% CI] = 0.024–0.81; p = 0.029), or tumor size, histological grade and patients' age (hazard ratio [HR] = 0.12, 95% confidence interval [95% CI] = 0.015–0.97; p = 0.047).
Fig. 3. Kaplan–Meier curves for (A) disease-free survival (DFS) and (B) overall survival (OS) of BC patients with tumors demonstrating high or low BCL2L12 expression.
Discussion It is well known that members of the BCL2 family are implicated in cancer pathogenesis and progression. Therefore, plenty of these apoptosis-related genes and proteins have been studied in many malignancies, including BC, for their prognostic and predictive potential [26]. For example, ERa-negative BC patients exhibit decreased expression of BAK, a member of the BCL2 member, suggesting that BAK down-
Fig. 2. Box plots representing BCL2L12 expression levels in relation to (A) disease stage and (B) to tumor size. The p value was calculated by the “Mann–Whitney U test” and “Jonckheere– Terpstra test”, respectively. The box refers to the range defined by the 25th and the 75th percentile and the bold line corresponds to the median value.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx Table 3 Cox univariate regression analysis of BCL2L12 regarding disease-free survival. Univariate analysis for disease-free survival ( N = 101) Variable BCL2L12 Low expression High expression Grade I/II III Size (Ordinal) Age (Ordinal) NPI (Ordinal) ER Status Negative Positive PR status Negative Positive HER-2 status (Ordinal)
HR
95% CI
p valuea
1.00 0.19
0.04–0.85
0.030
1.00 0.99 1.59 2.62 0.76
0.41–2.44 0.71–3.60 0.92–7.49 0.37–1.57
0.99 0.26 0.07 0.46
1.00 0.17
0.05–0.54
0.003
1.00 0.02 1.18
0.00–1.19 0.81–1.72
0.06 0.39
Abbreviations: HR = hazard ratio; CI = confidence intervals; NPI = Nottingham prognostic index. Statistically significant results are in bold type. a Test for trend.
regulation is associated with malignant development in the mammary gland [27]. On the other hand, BCL2, the prototypic member of the BCL2 family, appears to be a powerful independent favorable prognostic factor that can significantly influence the current prognostic models [20, 28–31], leading to its incorporation into the 21-gene panel of Oncotype DX [6]. The favorable prognostic nature of BCL2 in BC, is reinforced by its positive correlation with ER and PR expression [28,32,33], whereas loss
Table 4 Cox multivariate regression analysis of BCL2L12 regarding disease-free survival. Multivariate analysis for disease-free survival ( N = 89) Full modela HR
95% CI
p valueg
1.00 0.12
0.015–0.98
0.048
1.00 1.05 1.03 1.62 0.67
0.32–3.47 0.35 – 3.02 0.49–5.37 0.26–1.71
0.94 0.96 0.43 0.40
Variable
HR
95% CI
p value⁎
BCL2L12 Model 1b Model 2c Model 3d Model 4e Model 5f Model 6g
0.15 0.17 0.23 0.12 0.10 0.14
0.03–0.72 0.03–0.85 0.05–1.03 0.015–0.97 0.018–0.56 0.024–0.81
0.018 0.031 0.054 0.047 0.009 0.029
Variable BCL2L12 Low expression High expression Grade I/II III Size (Ordinal) Age (Ordinal) Nottingham Prognostic Index (Ordinal) Reduced models
Abbreviations: HR = hazard ratio; CI = confidence intervals; NPI = Nottingham prognostic index. Statistically significant results are in bold type. a Multivariate model adjusted for patients' age, tumor size, histological grade, and Nottingham prognostic index. b Multivariate model adjusted for ER status. c Multivariate model adjusted for PR status. d Multivariate model adjusted for HER-2 status. e Multivariate model adjusted for patients' age, tumor size and histological grade. f Multivariate model adjusted for patients' age, NPI and ER status. g Multivariate model adjusted for patients' age, NPI and PR status. ⁎ Test for trend.
5
of BCL2 expression is associated with the presence of unfavorable prognostic molecular markers (e.g. EGFR, HER2, and p53) [33]. BCL2L12 is a novel member of the BCL2 family, which is subjected to alternative splicing resulting in the generation of 13 spliced variants. Recently, the expression profile of eight novel BCL2L12 transcript variants was examined in several cancer cell lines and significant variations in their expression profile were observed. Regarding BC, BCL2L12 variants 4, 5, 10, 11 and 13 demonstrated differential expression among the MCF-7, BT474 and BT-20 cell lines [34]. The BCL2L12 mRNA expression has also been examined in tissue samples obtained from various malignancies, suggesting its possible prognostic role in cancer. As far as BC is concerned, patients bearing BCL2L12-positive tumors seem to have significantly longer DFS and OS [19,20]. In view of the above mentioned data we investigated the possible prognostic role of BCL2L12 in a large cohort of cancerous and non-cancerous breast specimens, using a highly sensitive qRT-PCR method. Based on our results, increased BCL2L12 expression levels were significantly associated with early TNM stages (p = 0.009), absence of metastasis (p = 0.012) and small tumor size (p = 0.04), supporting the previously reported hypothesis that mRNA expression of BCL2L12 is associated with less aggressive forms of BC [19,20]. Tumor size is an independent prognostic factor for BC and its increase is accompanied by decreased survival rates and high risk of recurrence [35–37]. The correlation of BCL2L12 expression with tumor size is interesting and may prove useful for the identification of women with favorable prognosis. Moreover, elevated levels of BCL2L12 mRNA were observed in patients 35–55 years old (p = 0.024). Epidemiological studies suggest an exponential increase of this malignancy until the age of 50 followed by a slight decrease around menopause age and then a linear rise, a phenomenon known as Clemmensen hook [38–40]. Many prospective and retrospective studies suggest that younger (≤35 years old) BC patients exhibit worse clinical outcome compared to the older ones (40–69 years old), having a significant higher recurrence risk [38,41–43]. Furthermore, although limited, there are data implying that age is an independent marker of poor prognosis in BC [43]. According to our results, women belonging to the age group characterized by low incidence rates and good prognosis, namely 35–55 years old, exhibited increased levels of BCL2L12 expression, enhancing the favorable nature of this gene. According to Kaplan–Meier (p = 0.015) and univariate Cox regression analysis (p = 0.030), there is a significant association of BCL2L12 overexpression with superior DFS intervals, supporting the favorable prognostic role of this gene. In more detail, patients with high BCL2L12 expression were almost five times less likely to relapse in relation to women with decreased BCL2L12 mRNA levels. Furthermore, multivariate regression analysis, adjusted for non-molecular parameters, namely, histological grade, tumor size, patients' age and Nottingham prognostic index, revealed that BCL2L12 expression can independently predict good recurrence-free survival (p = 0.048). In addition, a reduced model consisting of the same parameters, but excluding the Nottingham prognostic index, highlighted the significance of BCL2L12 mRNA expression as an independent predictor of DFS (p = 0.047). Since measurement of ER, PR and HER-2 is an established clinical practice for BC patients' prognosis, we deemed the incorporation of these markers in multiparametric logistic regression models necessary. A multivariate model including BCL2L12 expression profile and ER status, revealed that increased levels of BCL2L12 can independently predict a superior DFS (hazard ratio [HR] = 0.15; 95% confidence interval [95% CI] = 0.03–0.72; p = 0.018). Similar results were observed when multiparametric models were adjusted for PR status (hazard ratio [HR] = 0.17; 95% confidence interval [95% CI] = 0.03–0.85; p = 0.031) but not for HER-2 expression profile (hazard ratio [HR] = 0.23; 95% confidence interval [95% CI] = 0.05–1.03; p = 0.054). Finally, two additional reduced models adjusted for patients' age, NPI and ER (p = 0.009) or PR (p = 0.029) status revealed that BCL2L12 mRNA expression is a strong independent prognostic marker for BC.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
6
A. Tzovaras et al. / Clinical Biochemistry xxx (2014) xxx–xxx
Conclusion A possible prognostic role of BCL2L12 in BC has been revealed by previous experimental data generated from conventional PCR. In this study BCL2L12 expression was quantified in a large cohort of BC patients with a long-term follow up, using a sensitive qRT-PCR approach. According to our results, BCL2L12 is an independent favorable prognostic marker in BC, suggesting its use in the assessment of relapse risk in this type of cancer. However, the current study represents a first attempt at assessing the role of BCL2L12 as a potential tumor marker in BC and future studies, in an independent cohort, are needed for validation of our results and to further demonstrate its role in clinical practice. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2014.09.008. Conflict of Interest We declare that we have no conflict of interest Acknowledgements This research has been co-financed by the European Union (European Social Fund; ESF) and Greek national funds through the Operational Program “Education and Lifelong Learning” of the National Strategic Reference Framework (NSRF), Research Funding Program: THALES, Investing in Knowledge Society through the European Social Fund (THALES UoA-Biopromo, MIS 377046). References [1] Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011;61:69–90. [2] Polyak K. Heterogeneity in breast cancer. J Clin Invest 2011;121:3786–8. [3] Kittaneh M, Montero AJ, Gluck S. Molecular profiling for breast cancer: a comprehensive review. Biomark Cancer 2013;5:61–70. [4] Ellsworth RE, Decewicz DJ, Shriver CD, Ellsworth DL. Breast cancer in the personal genomics era. Curr Genomics May 2010;11(3):146–61. http://dx.doi.org/10.2174/ 138920210791110951. [5] Tian S, Roepman P, Van't Veer LJ, Bernards R, de Snoo F, Glas AM. Biological functions of the genes in the mammaprint breast cancer profile reflect the hallmarks of cancer. Biomark Insights 2010;5:129–38. [6] Cronin M, Sangli C, Liu ML, Pho M, Dutta D, Nguyen A, et al. Analytical validation of the oncotype dx genomic diagnostic test for recurrence prognosis and therapeutic response prediction in node-negative, estrogen receptor-positive breast cancer. Clin Chem 2007;53:1084–91. [7] Parker JS, Mullins M, Cheang MC, Leung S, Voduc D, Vickery T, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol Mar. 10 2009; 27(8):1160–7. http://dx.doi.org/10.1200/JCO.2008.18.1370 [Epub 2009 Feb 9. 2009]. [8] Scorilas A, Kyriakopoulou L, Yousef GM, Ashworth LK, Kwamie A, Diamandis EP. Molecular cloning, physical mapping, and expression analysis of a novel gene, bcl2l12, encoding a proline-rich protein with a highly conserved bh2 domain of the bcl-2 family. Genomics 2001;72:217–21. [9] Yousef GM, Kopolovic AD, Elliott MB, Diamandis EP. Genomic overview of serine proteases. Biochem Biophys Res Commun 2003;305(1):28–36. [10] Scorilas A, Chiang PM, Katsaros D, Yousef GM, Diamandis EP. Molecular characterization of a new gene, ceal1, encoding for a carcinoembryonic antigen-like protein with a highly conserved domain of eukaryotic translation initiation factors. Gene May 22 2003;310:79–89 [2003]. [11] Borglum AD, Byskov A, Ragno P, Roldan AL, Tripputi P, Cassani G, et al. Assignment of the urokinase-type plasminogen activator receptor gene (plaur) to chromosome 19q13.1-q13.2. Am J Hum Genet Mar. 1992;50(3):492–7 [1992]. [12] Florou D, Papadopoulos IN, Scorilas A. Molecular analysis and prognostic impact of the novel apoptotic gene bcl2l12 in gastric cancer. Biochem Biophys Res Commun 2010;391:214–8. [13] Foutadakis S, Avgeris M, Tokas T, Stravodimos K, Scorilas A. Increased bcl2l12 expression predicts the short-term relapse of patients with tat1 bladder cancer following transurethral resection of bladder tumors. Urol Oncol 2014;32:39.e29–36. [14] Fendri A, Kontos CK, Khabir A, Mokdad-Gargouri R, Scorilas A. Bcl2l12 is a novel biomarker for the prediction of short-term relapse in nasopharyngeal carcinoma. Mol Med 2011;17:163–71. [15] Geomela PA, Kontos CK, Yiotakis I, Scorilas A. Quantitative expression analysis of the apoptosis-related gene, bcl2l12, in head and neck squamous cell carcinoma. J Oral Pathol Med 2013;42:154–61.
[16] Kontos CK, Papadopoulos IN, Scorilas A. Quantitative expression analysis and prognostic significance of the novel apoptosis-related gene bcl2l12 in colon cancer. Biol Chem 2008;389:1467–75. [17] Thomadaki H, Floros KV, Pavlovic S, Tosic N, Gourgiotis D, Colovic M, et al. Overexpression of the novel member of the bcl2 gene family, bcl2l12, is associated with the disease outcome in patients with acute myeloid leukemia. Clin Biochem 2012; 45:1362–7. [18] Papageorgiou SG, Kontos CK, Pappa V, Thomadaki H, Kontsioti F, Dervenoulas J, et al. The novel member of the bcl2 gene family, bcl2l12, is substantially elevated in chronic lymphocytic leukemia patients, supporting its value as a significant biomarker. Oncologist 2011;16:1280–91. [19] Talieri M, Diamandis EP, Katsaros N, Gourgiotis D, Scorilas A. Expression of bcl2l12, a new member of apoptosis-related genes, in breast tumors. Thromb Haemost 2003; 89:1081–8. [20] Thomadaki H, Talieri M, Scorilas A. Prognostic value of the apoptosis related genes bcl2 and bcl2l12 in breast cancer. Cancer Lett 2007;247:48–55. [21] McShane LM, Altman DG, Sauerbrei W, Taube SE, Gion M, Clark GM. Reporting recommendations for tumor marker prognostic studies (remark). Breast Cancer Res Treat 2006;100:229–35. [22] Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative pcr and the 2(−delta delta c(t)) method. Methods Dec. 2001;25(4): 402–8 [2001]. [23] de Kok JB, Roelofs RW, Giesendorf BA, Pennings JL, Waas ET, Feuth T, et al. Normalization of gene expression measurements in tumor tissues: comparison of 13 endogenous control genes. Lab Invest Jan. 2005;85(1):154–9 [2005]. [24] Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology Apr. 1982;143(1):29–36 [1982]. [25] Camp RL, Dolled-Filhart M, Rimm DL. X-tile: a new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clin Cancer Res 2004;10: 7252–9. [26] Thomadaki H, Scorilas A. Bcl2 family of apoptosis-related genes: functions and clinical implications in cancer. Crit Rev Clin Lab Sci 2006;43:1–67. [27] Eguchi H, Suga K, Saji H, Toi M, Nakachi K, Hayashi SI. Different expression patterns of bcl-2 family genes in breast cancer by estrogen receptor status with special reference to pro-apoptotic bak gene. Cell Death Differ 2000;7:439–46. [28] Hwang KT, Woo JW, Shin HC, Kim HS, Ahn SK, Moon HG, et al. Prognostic influence of bcl2 expression in breast cancer. Int J Cancer 2012;131:E1109–19. [29] Dawson SJ, Makretsov N, Blows FM, Driver KE, Provenzano E, Le Quesne J, et al. Bcl2 in breast cancer: a favourable prognostic marker across molecular subtypes and independent of adjuvant therapy received. Br J Cancer 2010;103:668–75. [30] Parton M, Dowsett M, Smith I. Studies of apoptosis in breast cancer. BMJ 2001;322: 1528–32. [31] Sjostrom J, Blomqvist C, von Boguslawski K, Bengtsson NO, Mjaaland I, Malmstrom P, et al. The predictive value of bcl-2, bax, bcl-xl, bag-1, fas, and fasl for chemotherapy response in advanced breast cancer. Clin Cancer Res 2002;8:811–6. [32] Kolacinska A, Chalubinska J, Zawlik I, Szymanska B, Borowska-Garganisz E, Nowik M, et al. Apoptosis-, proliferation, immune function-, and drug resistance- related genes in ER positive, her2 positive and triple negative breast cancer. Neoplasma 2012;59: 424–32. [33] Hellemans P, van Dam PA, Weyler J, van Oosterom AT, Buytaert P, Van Marck E. Prognostic value of bcl-2 expression in invasive breast cancer. Br J Cancer 1995; 72:354–60. [34] Kontos CK, Scorilas A. Molecular cloning of novel alternatively spliced variants of bcl2l12, a new member of the bcl2 gene family, and their expression analysis in cancer cells. Gene Aug. 15 2012;505(1):153–66. http://dx.doi.org/10.1016/j.gene.2012. 04.084 [Epub 2012 Jun 1. 2012]. [35] Carter CL, Allen C, Henson DE. Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases. Cancer 1989;63:181–7. [36] Cianfrocca M, Goldstein LJ. Prognostic and predictive factors in early-stage breast cancer. Oncologist 2004;9:606–16. [37] Narod SA. Tumour size predicts long-term survival among women with lymph node-positive breast cancer. Curr Oncol 2012;19:249–53. [38] Nixon AJ, Neuberg D, Hayes DF, Gelman R, Connolly JL, Schnitt S, et al. Relationship of patient age to pathologic features of the tumor and prognosis for patients with stage I or II breast cancer. J Clin Oncol 1994;12:888–94. [39] Xiong Q, Valero V, Kau V, Kau SW, Taylor S, Smith TL, et al. Female patients with breast carcinoma age 30 years and younger have a poor prognosis: The M.D. Anderson Cancer Center experience. Cancer 2001;92:2523–8. [40] Kroman N, Jensen MB, Wohlfahrt J, Mouridsen HT, Andersen PK, Melbye M. Factors influencing the effect of age on prognosis in breast cancer: population based study. BMJ 2000;320:474–8. [41] Keating Litton J. Is age still the deciding factor? Commentary on “The effect of age on delay in diagnosis and stage of breast cancer” by Partridge et al. Oncologist 2012;17: 744–6. [42] Assi HA, Khoury KE, Dbouk H, Khalil LE, Mouhieddine TH, El Saghir NS. Epidemiology and prognosis of breast cancer in young women. J Thorac Dis 2013;5:S2–8. [43] Copson E, Eccles B, Maishman T, Gerty S, Stanton L, Cutress RI, et al. Prospective observational study of breast cancer treatment outcomes for UK women aged 18–40 years at diagnosis: the POSH Study. J Natl Cancer Inst 2013;105:978–88.
Please cite this article as: Tzovaras A, et al, BCL2L12: A promising molecular prognostic biomarker in breast cancer, Clin Biochem (2014), http:// dx.doi.org/10.1016/j.clinbiochem.2014.09.008
