l-DOPA Decarboxylase (DDC) Expression Status as a Novel Molecular Tumor Marker for Diagnostic and Prognostic Purposes in Laryngeal Cancer

Abstract

l-DOPA decarboxylase (DDC) plays an essential role in the enzymatic synthesis of dopamine and alterations in its gene expression have been reported in several malignancies. Our objective was to analyze DDC messenger RNA (mRNA) and protein expression in laryngeal tissues and to evaluate the clinical implication of this molecule in laryngeal cancer. In this study, total RNA was isolated from 157 tissue samples surgically removed from 100 laryngeal cancer patients. A highly sensitive real-time polymerase chain reaction methodology based on SYBR Green I fluorescent dye was developed for the quantification of DDC mRNA levels. In addition, Western blot analysis was performed for the detection of DDC protein. DDC mRNA expression was revealed to be significantly downregulated in primary laryngeal cancer samples compared with their nonmalignant counterparts (P = .001). A significant negative association was also disclosed between DDC mRNA levels and TNM staging (P = .034). Univariate analysis showed that patients bearing DDC-positive tumors had a significantly decreased risk of death (hazard ratio = 0.23, P = .012) and local recurrence (hazard ratio = 0.32, P =.006), whereas DDC expression retained its favorable prognostic significance in the multivariate analysis. Kaplan-Meier curves further demonstrated that DDC-positive patients experienced longer overall and disease-free survival periods (P = .006 and P = .004, respectively). Moreover, DDC protein was detected in both neoplastic and noncancerous tissues. Therefore, our results suggest that DDC expression status could qualify as a promising biomarker for the future clinical management of laryngeal cancer patients.

Introduction

Laryngeal cancer, the most frequent type of head and neck malignancy, accounts for approximately 2.2% of all cancer cases and 2.0% of all cancer-related deaths in the European male population [1]. Approximately 95% of all laryngeal tumors are histologically categorized as laryngeal squamous cell carcinomas (LSCCs), the development of which is mainly linked to tobacco smoking and excessive alcohol consumption, according to a plethora of epidemiological studies [2,3]. At present, the prediction of survival of patients with LSCC and the subsequent treatment selection rely on a number of tumor characteristics, such as the TNM stage, the anatomic site of the tumor, and the presence of lymph node metastases. Of the previously mentioned clinicopathologic factors, the nodal status holds a prominent place in the prognosis of this neoplasia because the presence of involved lymph nodes decreases the long-term survival by 50% [2,4,5]. However, none of these clinical parameters alone are adequate for accurately staging patients in proper prognostic groups in order for them to receive the most suitable treatment regimen.

During the last decade, the noteworthy progress in the translational research area, through considerable advances in molecular biology techniques and immunohistochemistry, in combination with the expansion of knowledge regarding the molecular networks underlying the pathogenesis and the development of laryngeal cancer, has been the stimulus for characterizing novel biologic markers with potential clinical applications in the diagnosis and the prognosis of the disease [4,6,7]. The candidate indicators for LSCC investigated so far participate in a diverse range of physiological functions, including apoptosis, cell cycle regulation, metabolism, immune response, and angiogenesis. Unfortunately, none of these biomarkers have been incorporated into the clinical management or treatment algorithms for laryngeal cancer patients as it is essential to further evaluate their proposed diagnostic and/or prognostic advantages [8,9]. Under these circumstances, it is of utmost importance to identify new molecular agents that may serve as diagnostic and/or prognostic markers to fuel up the effort against this fatal malignancy.

l-DOPA decarboxylase (DDC, EC 4.1.1.28) is a pyridoxal-5′-phosphate-dependent enzyme participating in the biosynthesis of catecholamines, a major class of neurotransmitters in the central nervous system, with the principal role of catalyzing the decarboxylation of l-3,4-dihydroxyphenylalanine (l-DOPA) to dopamine [10]. The enzyme is composed of two homologous subunits, each of which with a molecular weight of approximately 50 kDa and has been purified and characterized from human pheochromocytoma [11] and kidney [12]. DDC presents a wide distribution in human tissues; from the central nervous system, where it exerts its neurotransmitter biosynthetic function, to a series of peripheral organs, such as the liver, kidney, adrenals, lungs, gastrointestinal tract, and pancreas, in which its biologic purpose is yet to be determined [10,13]. Although it has been proposed that DDC may be implicated in the enzymatic conversion of other aromatic amino acids in the extraneuronal tissues [10], increasing evidence indicate that peripheral dopamine may be involved in cell survival and proliferation [14].

The gene coding for DDC consists of 15 exons, covering more than 85 kb of genomic DNA and is located to the chromosomal region 7p12.1–12.3 [15]. To date, a variety of different transcripts has been described as the result of alternative splicing events, taking place either in the 5′-untranslated or in the coding region of the DDC gene [16–18]. Alterations in the messenger RNA (mRNA) expression patterns of DDC have been associated with the diagnosis and prognosis of various human malignancies. Although DDC expression status is considered a well-established marker for neuroendocrine tumors [19,20], its clinical usefulness has been recently proved in other solid neoplasms as well [21,22].

To shed some light into the clinical significance of DDC expression and to evaluate its applicability as a prospective tissue biomarker for enhancing the diagnostic and prognostic effectiveness in laryngeal cancer, we analyzed the transcriptional activity of DDC in a set of LSCC tissues and assessed its relationship with clinicopathologic features and patient survival data by means of a highly specific and sensitive real-time quantitative polymerase chain reaction (qPCR) assay. Furthermore, to gain better insight about the protein product of DDC gene, we performed immunoblot analysis experiments in cancerous and normal laryngeal tissue extracts.

Materials and Methods

Laryngeal Cancer Patients and Tissue Specimens

For this retrospective study, 157 tissue samples were obtained from 100 patients with histologically and clinically diagnosed LSCC, who underwent surgical treatment for laryngeal cancer at Hippokration Hospital, Athens, Greece, during the period between 2005 and 2010. All resected tissue specimens, including 80 primary LSCC samples, 20 recurrent LSCC specimens, and 57 paired surrounding nonmalignant laryngeal tissues, were snap-frozen in liquid nitrogen immediately after their surgical excision and stored at -80°C before homogenization. None of the patients with primary LSCC received preoperatively radiotherapy or chemotherapy, whereas most patients who presented with local relapses were submitted to postsurgical radiotherapy sessions.

Demographic data and clinical information of each patient of the study cohort, such as age, sex, profession, tobacco and/or alcohol consumption habits, follow-up period, and postoperative treatment, were retrieved from the medical records of the hospital along with the histopathologic characteristics of tumors (anatomic location, size, TNM stage, and histologic grade). Of all the patients, 95% were men with a mean age of 63.2 years at the time of the surgery (SE = ±0.92), ranging from 36 to 90 years. Approximately all patients had been exposed to etiological factors of the disease such as tobacco smoke and/or alcohol abuse. Among 99 patients, for whom complete follow-up information concerning overall survival (OS) and disease-free survival (DFS) was available, 31 (31.3%) showed local recurrence and 18 (18.2%) died of LSCC. The follow-up period ranged from 4 to 60 months, with a median of 37 months for primary laryngeal cancer patients and 36 months for those diagnosed with recurrent disease.

All scientific manipulations were conducted in agreement with the ethical principles of the 1975 Declaration of Helsinki, as revised in 2000. In addition, the use of these clinical materials for research purposes and the experimental protocols applied in this study were approved by the Hippokration Hospital of Athens Ethics Committee, after notification of each patient participating in this study and acquisition of a signed informed consent form.

Cancer Cell Line and Culture Conditions

BT-474 is a cancer cell line of epithelial origin (number HTB-20; ATCC, Wesel, Germany), initially isolated from invasive breast ductal carcinoma. BT-474 cells were plated and cultured in RPMI 1640 growth medium (PAA, Pasching, Austria) supplemented with 10% fetal bovine serum, 100 U/ml penicillin, 100 µg/ml streptomycin, 0.3 g/ml l-glutamine, 0.85 g/L NaHCO₃, and 10 µg/ml insulin at 37°C in a humidified atmosphere supplied with 5% (vol/vol) CO₂.

Total RNA Isolation and Reverse Transcription

Total RNA was extracted from 10 to 100 mg of snap-frozen LSCC or tumor-adjacent normal tissue fragments, as well as from BT-474 cells, using TRI Reagent (Ambion, Austin, TX) in accordance with the manufacturer's protocol. The isolated RNA was preserved in RNA Storage solution (Ambion) and stored at -80°C. The concentration of the RNA solution was determined spectrophotometrically by measuring the absorbance at 260 nm and its purity was estimated on the basis of the absorbance ratio at 260 and 280 nm. Moreover, the integrity of the extracted RNA was evaluated by electrophoresis on 1.5% agarose gel stained with ethidium bromide.

From 1 µg of total cellular or laryngeal tissue RNA, 20 µl of first-strand complementary DNA (cDNA) was synthesized with the use of Oligo-dT primers and the M-MuLV Reverse Transcriptase RNase H^- (Finnzymes, Vantaa, Finland), combined with the Human Placental RNase Inhibitor (H T Biotechnology Ltd, Cambridge, United Kingdom). The resulting cDNA was used for subsequent real-time qPCR experiments.

Quantitative Real-time PCR

Quantification of DDC mRNA expression levels in the laryngeal tissue samples was achieved with the development of a real-time PCR methodology of exceptional sensitivity and reliability, based on the SYBR Green I fluorescent dye detection chemistry (Applied Biosystems, Foster City, CA). Two pairs of gene-specific oligonucleotide primers were designed and synthesized on the basis of the reference mRNA sequences published on the NCBI database, one set for the gene of interest, DDC (GenBank accession number NM_000790), and the other for the internal control gene of the method, GAPDH (GenBank accession number NM_002046). GAPDH, the gene that encodes glyceraldehyde3-phosphate dehydrogenase, was chosen as the reference gene because of the stability of its expression and its overall suitability in the normalization of quantitative expression data in the field of head and neck squamous cell carcinoma [23]. The sequences of the two sets of primers used are listed as follows: for GAPDH, the forward 5′-CCT CCC GCT TCG CTC TCT-3′ and the reverse 5′-CCG TTG ACT CCG ACC TTC AC-3′ primers produced an amplicon of 116 bp, whereas for DDC, the forward 5′-GAA CAG ACT TAA CGG GAG CCT TT-3′ and the reverse 5′-AAT GCC GGT AGT CAG TGA TAA GC-3′ gave rise to a 90 bp amplicon. The primers were designed to attach to separate exons to avoid amplification of contaminating genomic DNA.

Real-time qPCR for both genes was performed in triplicate using an ABI Prism 7500 thermal cycler (Applied Biosystems) for the whole set of laryngeal tissue samples. The reaction mixture contained 0.4 µlof cDNA, 50 nM of gene-specific primers and 2x Power SYBR Green PCR MasterMix (Applied Biosystems) in a final reaction volume of 10 µl. Thermal cycling included an initial 10-minute step at 95°C for the denaturation of the sample and the activation of AmpliTaq Gold hot-start DNA polymerase, and the mixture was then subjected to 40 cycles of denaturation at 95°C for 15 seconds and primer annealing and extension at 60°C for 1 minute. Following the amplification process, a dissociation curve analysis was performed to confirm the specificity of the qPCR and to verify the presence of the desired amplified sequences, by determining the characteristic melting temperature (T_m) of each generated PCR amplimer (Figure 1, B and C).

Figure 1.

Quantitative determination of DDC expression in laryngeal tissue samples by qPCR. Amplification plots (A) for DDC and GAPDH in a randomly selected laryngeal cancerous specimen, in which C_t values for both genes, as well as ΔC_t, are illustrated. Dissociation curves of the GAPDH (B) and DDC (C) PCR products, in which a single peak at 84.3 and 80.4°C, respectively, corresponding to the desired amplicon, is shown.

The mRNA expression of DDC for each laryngeal specimen was calculated with the comparative C_t method (2^-ΔΔCt), a relative quantification method enabling us to estimate relative changes in the transcriptional activity of the gene of interest [24], using the C_t values deduced from the real-time PCR fluorescent emission data and a calibrator, which in our experiments was the human breast tumor cell line BT-474. Hence, the comparison of DDC expression, normalized to GAPDH, between different laryngeal tissue samples in the study was feasible by setting the calibrator's normalized DDC expression as a reference point. Our results were interpreted by introducing the arbitrary relative quantification (RQ) unit, calculated by the arithmetic formula 2^-ΔΔCt, where ΔΔC_t equals to [(C_{t DDC} - C_{t GAPDH})_sample - (C_{t DDC} - C_{t GAPDH}) _calibrator] for every given sample. The RQ unit, therefore, represents the normalized-to-GAPDH amounts of DDC mRNA levels of a laryngeal tissue sample relative to the normalized DDC expression of BT-474 cancer cells and provides a useful way to estimate the relative changes in DDC expression between different samples.

Immunodetection of DDC Protein

A small subset of eight randomly selected samples, four LSCCs and four matched nonneoplastic laryngeal tissues, whose DDC expression at the mRNA level was recorded as positive in the qPCR experiments, were further analyzed by Western blot. Total protein was extracted from the laryngeal samples according to the TRI Reagent protein isolation protocol (Ambion) and dissolved in 1% sodium dodecyl sulfate (SDS). The total protein concentration of each sample was spectrophotometrically determined at 595 nm, based on the Bradford protein assay with bovine serum albumin as standard.

Protein samples (5 µg/lane) were separated at room temperature by 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) on an electrophoresis device (Cleaver Scientific, Rugby, United Kingdom) at 150 V for 90 minutes and then electrophoretically transferred onto a nitrocellulose membrane (Schleicher & Schuell Bioscience, Dassel, Germany) using Trans-Blot SD semidry cell (Bio-Rad, Hemel Hempstead, United Kingdom). After blocking overnight at 4°C in 1x TNT (10 mM Tris, 150 mM NaCl, 0.05% Tween 20) with 10% bovine serum albumin, DDC immunodetection was carried out using a polyclonal rabbit primary antibody (diluted 1:1000) against 32 amino acids of the C-terminal region of human DDC, raised from our laboratory. The membranes were subsequently submitted to chromogenic staining with an alkaline phosphatase-conjugated goat anti-rabbit secondary antibody (diluted 1:20,000; Chemicon International, Temecula, CA), leading to the development of a characteristic band that corresponds to DDC.

Statistical Analysis

The nonparametric Wilcoxon signed ranks test was used for the analysis of the differences in DDC expression patterns between neoplastic and noncancerous laryngeal samples because the expression levels of DDC in these groups did not display a normal distribution. By using the X-tile algorithm, an optimal cutoff point for DDC was generated, equal to 0.06 RQ units, to classify DDC mRNA expression of each sample as positive or negative, depending on whether the RQ units were above or below this cutoff point. Relationships between DDC expression status and the clinicopathologic variables and patients' age were also examined by implementing the χ² test or Fisher exact test, where appropriate. The diagnostic assistance of DDC expression profiles was assessed by constructing receiver operating characteristic (ROC) curves, depicting the sensitivity versus (1 - specificity) for all possible cutoff values. The areas under the curves (AUCs) were analyzed using the method of Hanley and McNeil. Univariate and multivariate analyses were carried out using Cox proportional hazards models. Survival curves for OS and DFS were designed with the Kaplan-Meier method and compared using the log-rank test. The significance level was set at P < .05.

Results

Validation of DDC mRNA Relative Quantification in Laryngeal Tissues with qPCR

The quantitative determination of DDC mRNA expression in the complete collection of laryngeal tissue specimens was accomplished by selecting a relative quantification approach, the comparative C_t method (2^-ΔΔCt), with the intention of analyzing the acquired data from the two-step reverse transcription real-time PCR methodology implemented in this study. The comparative C_t method (2^-ΔΔCt), the principle of which requires a reference gene and a calibrator sample, takes advantage of the C_t values generated from the amplification plots of GAPDH and DDC (Figure 1A).

The basic assumption, however, to apply this relative quantification method, is that the amplification efficiencies of both target and reference genes must be approximately equal and, if possible, to exceed 90% [24]. Serially diluted samples of BT-474 cDNA, derived from 200 to 0.01 ng of total RNA, were subjected to amplification for both genes in an attempt to confirm the above hypothesis. The qPCR efficiencies were calculated from the slope of the standard curves created by plotting the C_t values against the logarithm of the input RNA mass (Figure 2). Conducting this experiment, comparable efficiencies for both GAPDH (3.5244, 92.2% efficiency, R² = 0.9939) and DDC (3.3745, 97.8% efficiency, R² = 0.9951) were demonstrated, thus rendering the 2^-ΔΔCt method acceptable for the quantitative evaluation of DDC transcriptional levels.

Figure 2.

Validation of the 2^-ΔΔCt method by constructing standard curves for both DDC and GAPDH using as template serial dilutions of BT-474 cDNA.

Analysis of DDC Expression in Laryngeal Tissues

The distribution of DDC mRNA expression pattern in patients with laryngeal cancer is displayed in Table 1. Detectable DDC transcript levels were observed in both cancerous and tumor-adjacent nonmalignant laryngeal tissues. Nevertheless, the expression analysis of DDC at the mRNA level showed a three-fold decrease in both primary (mean ± SE = 28.7 ± 9.0 RQ units) and recurrent (mean ± SE = 32.9 ± 16.7 RQ units) laryngeal carcinomas in comparison with the nonneoplastic laryngeal mucosa samples (mean ± SE = 98.1 ± 31.4 RQ units). No differentiation regarding DDC expression is noticed between primary and relapsed LSCCs. When the laryngeal specimens were examined as pairs of cancerous and matched noncancerous tissues, a statistically significant down-regulation of the DDC transcriptional activity (P = .001) was revealed only in the case of primary LSCCs, where approximately 78% of the patients demonstrated higher DDC mRNA levels in the surrounding nonmalignant tissue samples compared to their tumor counterparts (Table 2).

Table 1.

Descriptive Statistics of DDC Expression in Laryngeal Tumor Samples.

DDC Expression (RQ Units)	Mean ± SE*	Range	Quartiles
			Lower	Median	Upper
Noncancerous tissues (n = 57)	98.1 ± 31.4	0.03–1196.8	4.8	21.8	52.5
Primary cancerous tissues (n = 80)	28.7 ± 9.0	0.03–552.6	0.06	1.8	16.2
Recurrent cancerous tissues (n = 20)	32.9 ± 16.7	0.03–269.8	0.03	1.6	9.4

^*SE indicates standard error.

Table 2.

DDC Expression in Pairs of Cancerous and Noncancerous Laryngeal Tissue Specimens.

DDC Expression in Pairs	No. Patients (%)	P*
Primary tumors
Higher in noncancerous vs cancerous samples	35 (77.8)	.001
Lower in noncancerous vs cancerous samples	7 (15.6)
Equal in noncancerous and cancerous samples	3 (6.6)
Recurrent tumors
Higher in noncancerous vs cancerous samples	7 (63.6)	.37
Lower in noncancerous vs cancerous samples	4 (36.4)

^*Calculated using the nonparametric Wilcoxon signed ranks test.

The expression of DDC protein was screened in eight randomly selected laryngeal samples by Western blot analysis, using a human kidney cancer sample as a positive control. The immunoblot analysis experiment uncovered the existence of a singe-specific band for DDC at the anticipated molecular mass in both LSCC and tumor-free tissue specimens (Figure 3). In addition, it is clearly denoted that DDC protein is differentially expressed between the malignant and nonmalignant laryngeal samples analyzed, with the noncancerous ones to be characterized by greater DDC protein production. However, the DDC protein levels of these laryngeal tissues are consistent with their respective mRNA levels, as determined with real-time PCR.

Figure 3.

Immunologic detection of DDC in representative LSCCs and nonmalignant laryngeal tissues by using anti-DDC polyclonal antibody. Lanes refer to the following: 1, molecular weight marker; 2, positive control (human kidney cancer); 3, 5, 7, 9, nonmalignant laryngeal samples; 4, 6, 8, 10, tumor specimens.

Assessment of the Diagnostic Ability of DDC Expression Status in Laryngeal Cancer

ROC curve analysis was used to explore the discriminatory potential of DDC transcription levels between cancer of the larynx, of primary origin or recurrence, and the nonpathologic state (Figure 4). The exported ROC curves indicated the significant diagnostic value of DDC expression levels in LSCC (AUC = 0.71, 95% confidence interval = 0.63–0.81, P < .001 for primary tumors; and AUC = 0.73, 95% confidence interval = 0.59–0.87, P = .002 for recurrent ones) and support its role as a diagnostic classifier in laryngeal cancer.

Figure 4.

ROC curve analysis for DDC mRNA expression in laryngeal tissues demonstrating its potential role in the diagnosis of primary (A) or recurrent (B) LSCC.

Association of DDC Expression with Clinicopathologic Variables in Primary LSCC

In the next step of the study, we evaluated the possible correlations between DDC mRNA expression profile and a series of clinical and histopathologic parameters concerning laryngeal cancer patients with primary tumors, as synopsized in Table 3. Consistent with our results, a statistically significant inverse association (P = .034) was discovered between DDC expression status and the TNM stage of the disease. In particular, the frequency of DDC positivity was significantly lower in tumors of advanced TNM stage (II–IV), as opposed to those of stage I. However, no statistically significant relationships were established between DDC expression pattern and histologic grade (P = .99) or patients' age (P =.44).

Table 3.

Correlation between DDC Expression Status and Certain Clinicopathologic Features in Patients with Primary Laryngeal Cancer.

Variable	Total No. Patients	No. Patients (%)		P
		DDC-Negative*	DDC-Positive*
TNM stage
I	13	0 (0.0)	13 (100.0)	.034†
II–IV	67	18 (26.9)	49 (73.1)
Grade
1/2	44	10 (22.7)	34 (77.3)	.99†
3	32	7 (21.9)	25 (78.1)
x	4
Age (years)
<55	12	1 (8.3)	11 (91.7)	.44‡
55–65	36	9 (25.0)	27 (75.0)
>65	32	8 (25.0)	24 (75.0)

x indicates unknown.

^*Cutoff point = 0.06 RQ units (23rd percentile).

^†Calculated with the Fisher exact test.

^‡Calculated with the χ² test.

Prognostic Implications of DDC mRNA Expression in Primary Laryngeal Cancer

The predictive strength of DDC transcriptional status concerning the clinical course of patients coping with laryngeal cancer was specified by developing univariate and multivariate Cox proportional hazards regression models (Table 4). As disclosed by the univariate analysis, DDC expression, at the mRNA level, presented a statistically significant association with both OS (P = .012) and DFS (P =.006), given that patients with DDC-positive laryngeal carcinomas showed a decreased risk of death (hazard ratio [HR] = 0.23) or relapse (HR = 0.32). More than that, when other clinicopathologic variables were introduced in the multivariate analysis, the prognostication power of DDC mRNA profile remained, implying that DDC expression may serve as an independent favorable prognostic indicator for both OS (P = .008) and DFS (P = .002). As expected, Kaplan-Meier survival curves (Figure 5) also illustrated that patients harboring DDC-positive primary laryngeal tumors exhibited a significantly longer OS and DFS (P = .006 and P = .004, respectively), compared to those whose neoplasms were classified as DDC-negative.

Table 4.

Univariate and Multivariate Analyses of DDC Expression Levels regarding DFS and OS of Patients with Primary Laryngeal Cancer.

	DFS			OS
	HR*	95% CI†	P‡	HR*	95% CI†	P‡
Univariate analysis
DDC-negative	1.00			1.00
DDC-positive	0.32	0.14–0.73	.006	0.23	0.076–0.73	.012
Multivariate analysis§
DDC-negative	1.00			1.00
DDC-positive	0.23	0.09–0.57	.002	0.18	0.05–0.64	.008

^*Hazard ratios (HRs) were estimated from the Cox proportional hazard regression model.

^†Confidence interval of the estimated HR.

^‡Test for trend.

^§Multivariate models were adjusted for patients' age, tumor grade, and smoking and alcohol consumption.

Figure 5.

Kaplan-Meier survival curves for the DFS (A) and OS (B) of patients with primary laryngeal tumors categorized as either DDC-positive or DDC-negative.

Discussion

Cancer of the larynx constitutes the most prevalent form of head and neck malignancy in the Mediterranean basin, especially in Greece, where it is estimated to be responsible for more than 50% of these occurrences [25,26]. A recent meta-analysis, based on a series of published studies, revealed that the overall 5-year relative survival rate of laryngeal cancer patients was approximately 64%, one of the lowest comparing with other types of cancer [27], regardless of the latest advances in surgical procedures, combined treatment modalities and chemotherapeutic protocols that have been documented the past two decades [28]. Among the reasons behind this adverse prognosis of patients with LSCC is the clinically silent nature of this neoplasia at the time of presentation, the insufficiency of the TNM staging system to accurately stratify patients into more consistent prognostic groups, and the pronounced biologic heterogeneity of this disease, in which diverse molecular and genetic abnormalities are masked under seemingly identical phenotypes [8,28]. The major challenge, therefore, in laryngeal cancer is to escalate research efforts by focusing on the identification and development of new molecular-based tumor markers that could empower the clinical management and eventually expand the life span of patients with LSCC.

In the last few years, accumulated evidence has shown that dopa-mine, as well as other biogenic amines, participates in a wide range of biochemical pathways in the nonneuronal tissues, which directly relate to cancer progression [14]. One such physiological process, in which dopamine has been reported to mediate, is the formation of new blood vessels. It is now generally acknowledged that dopa-mine is an endogenous inhibitor of tumor angiogenesis, suppressing in this way the growth and proliferation of malignant tumors [29]. Another example of the antiproliferative effects of dopamine lies in gastric cancer, where it was recently demonstrated that dopamine negatively regulates tumor cell proliferation through specific molecular interactions resulting in cell cycle arrest [30]. Moreover, dopa-mine was also found to stimulate apoptosis in one leukemia and several oral squamous cell carcinoma cell lines [31,32]. Taking all the above into consideration, we investigated, in this research work, both the mRNA and protein expression of DDC, the enzyme responsible for the enzymatic conversion of l-DOPA to dopamine, in laryngeal tissue samples with the aim of elucidating its clinical utilization as a molecular biomarker for LSCC.

As far as we know, this is the first time the transcriptional and translational expression of DDC is being examined in laryngeal tissues. Apart from its existence in all aminergic cells of the central and peripheral neurotransmitter systems, DDC has also been detected in peripheral organs including liver, kidney, and pancreas [10]. Furthermore, Kokkinou et al. [33] have recently reported that enzymatically active DDC is localized in human peripheral leukocytes, thus expanding the already-accepted distribution pattern of this catecholamine biosynthetic enzyme.

Our study revealed detectable levels of DDC mRNA in all laryngeal tissue groups. However, a statistically significant underexpression of the gene (P = .001) in the primary LSCC tissue specimens was observed, when compared to the neighboring nonmalignant laryngeal samples. This discernible down-regulation of DDC in laryngeal cancer could be possibly attributed to the general deregulation of the cell's transcription mechanism to which it is subjected during the malignant transformation. The diagnostic utility of this particular finding in laryngeal cancer was highlighted by the ROC curve analysis, according to which the DDC mRNA expression profile could assist the diagnosis of LSCC, primary (AUC = 0.71, P < .001) or recurrent (AUC = 0.73, P =.002), by discriminating the noncancerous condition from the neoplastic disease. Previous studies have already proposed DDC transcriptional status as a propitious marker for the diagnosis of various malignancies. Specifically, Gilbert et al. [34] suggested that the levels of DDC expression could be exploited for the accurate differential diagnosis of neuroblastoma from other pediatric small round cell neoplasms, whereas it was also recommended as an indicator for detecting neuroblastoma cells in the bloodstream or the bone marrow with great specificity [19]. Similarly, the experiments by Avgeris et al. [21] supported the notion that DDC transcriptional modulations could be useful for distinguishing between prostate cancer and benign prostate hyperplasia. Consequently, we could safely assume that the fluctuations of DDC transcriptional activity in laryngeal cancer might contribute to the early diagnosis of this, often asymptomatic at the early stages, malignancy and the detection of minimal residual disease at the margins of the surgical resection, which is associated with increased risk of local recurrence and death.

Another interesting finding of our present research was the revelation of a statistically significant negative correlation between the mRNA expression of DDC and the TNM classification of the primary laryngeal tumors (P = .034). Of note, all stage I laryngeal carcinomas were DDC-positive, in contrast with the laryngeal tumors of advanced clinical stage (II-IV), from which a percentage of 73.1% exhibited expression levels above the calculated cutoff for DDC. Because early-stage laryngeal malignancies tend to present higher cure rates and usually the patients' life span could be prolonged, elevated DDC expression levels could comprise a molecular indicator of favorable prognosis for patients with LSCC. However, this is not the sole survey unveiling a fundamental relationship between DDC mRNA expression and the progression of different types of cancer. In the case of colorectal adenocarcinoma, the expression status of DDC was found to be significantly associated with the histologic grade of the tumors examined [22], whereas in patients with gastric cancer, DDC transcriptional status correlated with various features characterizing the natural course of this neoplasia, such as differentiation, depth of invasion, lymphatic invasion, and peritoneal dissemination [35]. Likewise, from a research work conducted in prostate cancer, a possible connection between DDC mRNA levels and the pathologic stage of this malignancy was emerged, implying the role of this enzyme in the development and progression of this neoplasia [36].

Along this line lies the fact that patients bearing DDC-positive primary laryngeal tumors live considerably longer (P =.006) and without disease relapses (P = .004) than those with LSCCs negative for DDC expression, as depicted on the Kaplan-Meier graphs. As estimated by the survival analysis, the 5-year OS for DDC-positive patients was 90% contrary to the approximately 62% for patients with DDC-negative laryngeal cancer. Comparable results were obtained from Cox univariate regression analysis, which clearly showed that patients coping with DDC-negative laryngeal neoplasms had about three times higher risk of experiencing locoregional recurrence (P = .006, HR = 0.32) and about four times greater probability of succumbing to the disease (P = .012, HR = 0.23). On top of that, the multivariate analysis stressed out the independent prognostic significance of DDC mRNA expression status as a favorable predictive marker for both OS (P =.008) and DFS (P = .002) of patients with primary LSCC. As an immediate consequence, DDC expression may act as a substantial and reliable indicator, alone or as component of a broader clinicomolecular multiparametric panel, for designating patients with LSCC into rigorous risk groups, so as to facilitate the selection of the appropriate candidates, who would benefit from more aggressive treatment approaches. Analogous observations were made by Kontos et al. [22] in colorectal adenocarcinoma, where it was described that the reduction in DDC mRNA amounts adversely influenced the clinical outcome of patients, pertaining not only the DFS but also the OS. Particularly, it was shown that patients with DDC-positive adenocarcinomas presented approximately five times higher survival probability, as well as about four times lower risk of recurrence. We could surmise that the prognostic impact of enhanced DDC transcriptional activity in the survival of LSCC patients might be associated with the accompanied increment in the production of dopamine, which, as mentioned before, has proved to possess distinct tumor-protective effects in numerous reports [14].

Adding to the clinical importance of DDC gene in laryngeal cancer, we concentrated our attention on the detection of its protein product in the laryngeal samples. According to Western blot analysis, a unique band representing DDC was successfully detected in both malignant and noncancerous laryngeal protein extracts, indicating that the DDC enzyme is normally expressed not only in the human larynx, perhaps as an integral part of the laryngeal physiology, but also in the neoplastic conditions of this organ. Recently, a study performed by Goulioumis et al. [37] suggested a critical role for androgen receptors (ARs) in the pathobiology of squamous laryngeal cancer, based on their immunohistochemical data regarding the expression of ARs in this neoplasia. Other authors evidenced that DDC interacts with AR at the cytoplasmic level, escalating its transcriptional activity, which, in turn, differentially modifies the expression patterns of AR-regulated genes in prostate cancer cells [38,39]. Hence, a reasonable speculation, arising from these surveys, is that DDC is implicated in the development and progression of laryngeal cancer through its ability to indirectly alter the mRNA levels of genes that are under the transcriptional control of ARs.

In conclusion, this is the first report in the literature that provides compelling evidence about the transcriptional and translational status of DDC in laryngeal tissues and eventually confirms the paramount diagnostic and prognostic opportunities emerging from the quantification of DDC mRNA expression levels in LSCC. Although future studies are indispensable to evaluate and integrate these findings into clinical decision making, the present research strongly signifies DDC as a promising molecular tumor marker for this head and neck malignancy.