PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of neoplasiaGuide for AuthorsAbout this journalExplore this journalNeoplasia (New York, N.Y.)
 
Neoplasia. 2010 October; 12(10): 807–817.
PMCID: PMC2950330

Gene Expression Levels as Predictive Markers of Outcome in Pancreatic Cancer after Gemcitabine-Based Adjuvant Chemotherapy1,2

Abstract

Background and Aims

The standard palliative chemotherapy for pancreatic ductal adenocarcinoma (PDAC) is gemcitabine-based chemotherapy; however, PDAC still presents a major therapeutic challenge. The aims of this study were to investigate the expression pattern of genes involved in gemcitabine sensitivity in resected PDAC tissues and to determine correlations of gene expression with treatment outcome.

Materials and Methods

We obtained formalin-fixed paraffin-embedded (FFPE) tissue samples from 70 patients with PDAC. Of the 70 patients, 40 received gemcitabine-based adjuvant chemotherapy (AC). We measured hENT1, dCK, CDA, RRM1, and RRM2 messenger RNA (mRNA) levels by quantitative real-time reverse transcription-polymerase chain reaction and determined the combined score (GEM score), based on the expression levels of hENT1, dCK, RRM1, and RRM2, to investigate the association with survival time. By determining the expression levels of these genes in endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) cytologic specimens, we investigated the feasibility of individualized chemotherapy.

Results

High dCK (P = .0067), low RRM2 (P = .003), and high GEM score (P = .0003) groups had a significantly longer disease-free survival in the gemcitabine-treated group. A low GEM score (<2) was an independent predictive marker for poor outcome to gemcitabine-based AC as shown by multivariate analysis (P = .0081). Altered expression levels of these genes were distinguishable in microdissected neoplastic cells from EUS-FNA cytologic specimens.

Conclusions

Quantitative analyses of hENT1, dCK, RRM1, and RRM2 mRNA levels using FFPE tissue samples and microdissected neoplastic cells from EUS-FNA cytologic specimens may be useful in predicting the gemcitabine sensitivity of patients with PDAC.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal and aggressive human malignancies, being the fourth leading cause of tumor-related deaths in the industrialized world [1,2]. Most patients with PDAC have poor outcomes because of the aggressive biology of the tumor and the difficulties of early diagnosis because of a lack of early disease-specific signs and symptoms. Only 10% to 20% of patients with PDAC are candidates for curative resection [3], and even if the curative resection is performed, the postoperative 5-year survival rate is only 15% to 25% because of a high recurrence rate [4,5]. Although two recent randomized clinical phase 3 trials of adjuvant chemotherapy (AC) for PDAC showed significant increases in overall survival (OS) and diseasefree survival (DFS) [6,7], there remain a substantial subset of cases in which AC efficacy is limited and insufficient. Recent studies have revealed that altered gene expression can at least partly explain responses and toxicity of cytotoxic agents [8]. To improve the prognosis of patients with PDAC, a helpful strategy would be to select subjects who are likely to respond to treatment based on gene expression profiles of the individual's own cancer tissues.

Gemcitabine (difluorodeoxycytidine; dFdC) is a deoxycytidine analog that has broad antitumor activity in various solid tumors, including pancreatic cancer [7] and non-small cell lung cancer (NSCLC) [9]. The drug is a prodrug that requires cellular uptake and intracellular phosphorylation to produce active diphosphate (dFdCDP) and triphosphate (dFdCTP). These phosphorylated forms function by inhibiting ribonucleotide reductase (RR) and DNA synthesis [10]. Gemcitabine is transported into cells predominantly by human equilibrative nucleoside transporter 1 (hENT1) [11]. A deficiency in hENT1 activity conferred high-level resistance to the toxicity of gemcitabine [12], and patients with PDAC that have detectable hENT1 or high hENT1 gene expression have significantly prolonged survival after gemcitabine chemotherapy [13,14]. After cellular entry, gemcitabine must be phosphorylated by deoxycytidine kinase (dCK), which is a rate-limiting step. We previously demonstrated that down-regulation of dCK specifically enhanced acquired resistance to gemcitabine in pancreatic cancer cells [15], whereas transfection of wild-type dCK restored sensitivity to the drug [16]. Conversely, active metabolites of gemcitabine are reduced by 5′-nucleotidase (5′-NT), and gemcitabine itself is inactivated by cytidine deaminase (CDA). High levels of these catabolic enzymes are associated with resistance to the drug [17,18]. dFdCTP inhibits DNA synthesis by being incorporated into the DNA strand, but in addition, dFdCDP potently inhibits RR, resulting in a decrease of competing deoxyribonucleotide pools necessary for DNA synthesis [19]. RR is a dimeric enzyme composed of regulatory subunit M1 and catalytic subunit M2. Recurrent PDAC patients with high levels of RRM1 expression had poor survival rates after gemcitabine treatment [20], and NSCLC patients with low levels of RRM1 expression significantly benefited from gemcitabine/cisplatin neoadjuvant chemotherapy [21]. Moreover, RRM2 gene silencing by RNA interference is an effective therapeutic adjunct to gemcitabine treatment [22]. These data suggest that the genes encoding proteins involved in the transport and metabolism of gemcitabine and in the metabolism of targets can be potential candidates to predict sensitivity to gemcitabine.

To develop individualized chemotherapy, the characterization of genes associated with tumor sensitivity or resistance to antitumor agents using cancer tissues from individuals plays a critical role in the selection of preferable treatments. In the current study, we investigated the correlation between the expression of genes involved in cellular uptake and metabolism of gemcitabine and the treatment outcome of patients with PDAC who underwent gemcitabine-based AC or no AC. Furthermore, to investigate the feasibility of individualized chemotherapy for patients with PDAC, even when the tumor is unresectable, we quantified the expression of genes in cytologic specimens obtained from endoscopic ultrasound-guided fine needle aspiration (EUS-FNA).

Materials and Methods

Cell Lines and Establishment of Gemcitabine-Resistant Cells

We used two pancreatic cancer cell lines, SUIT-2 (generously provided by Dr H. Iguchi, National Shikoku Cancer Center, Matsuyama, Japan) and Capan-1 (American Type Culture Collection, Manassas, VA). Gemcitabine-resistant Capan-1-GR and SUIT-2-GR cells were generated by exposing to gradually increasing concentrations of gemcitabine as described previously [15]. Cells were maintained as described previously [23].

Propidium Iodide Assay

To calculate the 50% inhibitory concentration (IC50) of each cell line when exposed to gemcitabine, cells were seeded in 24-well plates (Becton Dickinson Labware, Bedford, MA) at a density of 2 x 104 per well, using cell numbers previously counted using a particle distribution analyzer CDA 500 (Sysmex, Kobe, Japan). Several different concentrations of gemcitabine (Wako, Osaka, Japan) were added 24 hours after seeding. Cell populations were evaluated by measuring the fluorescence intensity of propidium iodide after a further incubation for 72 hours, as described previously [24].

Patients and Pancreatic Tissues

Our study subjects consisted of 70 patients including 40 patients who received gemcitabine-based AC (GEM group) and 30 patients who received no AC (non-AC group) after pancreatic resection for PDAC at the Department of Surgery and Oncology, Kyushu University Hospital (Fukuoka, Japan) from 1992 to 2007. Although there were 48 patients who received gemcitabine-based AC, eight patients were excluded because they did not receive adequate AC. The GEM group patients (n = 40) received gemcitabine-based AC, consisting of two or more cycles of 1000 mg/m2 per day of gemcitabine on days 1, 8, and 15 every 28 days, or three or more cycles of 1000 mg/m2 per day of gemcitabine on days 1 and 8 every 21 days. The patients were 42 men and 28 women with a median age of 65 years (range, 36–86 years). We recommended that patients have follow-up visits every 3 months for 2 years, then visits every 6 months for 3 years, and then annual visits. DFS was defined as the time from the date of pancreatic resection to the date of local or distant recurrence. The date of recurrence was defined as the date of the first subjective symptom heralding relapse, or the date of documentation of recurrent disease, independent of site, as assessed by diagnostic imaging techniques (whichever occurred first). Data for patients without recurrence were censored at the time of the last follow-up visit. OS was measured from the date of pancreatic resection to the date of death. Fifty-seven patients died during follow-up, and the other patients were censored at the time of the last follow-up visit. Data were analyzed in December 2009, and follow-up data from all cases were available. The median observation time for DFS was 8.0 months (range, 0.5–114 months) and that for OS was 15.7 months (range, 0.5–114 months). The clinicopathologic characteristics of the tumors collected from a total of 70 patients are noted in Table W1.

All resected specimens were fixed in formalin and embedded in paraffin for pathologic diagnosis. All tissues adjacent to the specimens were evaluated histologically according to the criteria of the World Health Organization. Two pathologists were in agreement with regard to the pathologic features of all cases and both confirmed the diagnoses. The stage of tumors was assessed according to UICC (Union Internationale Contre le Cancer and the American Joint Committee on Cancer) guidelines [25]. The study was approved by the ethics committee of Kyushu University and conducted according to the Ethical Guidelines for Human Genome/Gene Research enacted by the Japanese Government and the Helsinki Declaration.

Cytologic Specimens

Cytologic specimens were obtained at the time of cytologic examination and diagnosis from the pathologic laboratory of Kyushu University Hospital (Fukuoka, Japan). In brief, cytologic specimens were divided into whole cell pellets (WCPs) and into three or more smears as soon as possible after retrieval. Smears were processed in three different ways as described previously [26]. Two smears were mounted on standard glass slides for Hemacolor staining (Merck KGaA, Darmstadt, Germany) and Papanicolaou staining and were then used for rapid cytologic diagnosis and strict cytologic diagnosis, respectively. These two smears were examined histologically by cytopathologists, and diagnosis was confirmed according to Papanicolaou classification. The third smear of each specimen was mounted on membrane slides (P.A.L.M. Microlaser Technologies, Bernried, Germany) for laser capture microdissection. These smears were stained in 1% Toluidine blue staining solution or by Hemacolor staining. Fifteen cytologic specimens were obtained from patients at the Kyushu University Hospital (Fukuoka, Japan) who underwent endoscopic ultrasound-guided fine needle aspiration (EUS-FNA) cytology and whose lesions were cytopathologically diagnosed as PDAC.

Isolation of RNA

Total RNA was extracted from cultured cells using a High-Pure RNA Isolation Kit (Roche Diagnostics, Mannheim, Germany) according to the manufacturer's instructions. Total RNA was isolated from formalin-fixed paraffin-embedded (FFPE) tissue samples using the RNeasy FFPE kit (Qiagen, Tokyo, Japan) with modification of the manufacturer's instructions after macrodissection based on a review of representative hematoxylin and eosin-stained slides as described previously [27]. Total RNAwas extracted from cells isolated by microdissection according to the standard acid guanidinium thiocyanatephenol-chloroform (AGPC) protocol [28], with or without glycogen (Funakoshi, Tokyo, Japan).

Quantitative Real-time Reverse Transcription-Polymerase Chain Reaction

Quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR) was performed using a Chromo4 Real-time PCR Detection System (BIO-LAD Laboratories, Hercules, CA) and a LightCycler 480 II Real-time PCR System (Roche Diagnostics) for 40 cycles for 15 seconds at 95°C and 1 minute at 55°C with a Quanti-Tect SYBR Green Reverse Transcription-PCR kit (Qiagen) according to the manufacturer's instructions [29]. We designed specific primers (Table 1) and performed BLASTN searches to confirm the primer specificities. The level of each mRNA was calculated from a standard curve constructed with total RNA from Capan-1, a human pancreatic cancer cell line. The level of each mRNA was normalized to that of β-actin. The PCR product sizes of each primer pair are small, which allowed accurate and sensitive qRT-PCR despite the fragmented RNA extracted from FFPE tissue specimens [30,31].

Table 1
Primer Sequences and Product Sizes.

Statistical Analyses

Statistical analyses and graph presentations were made using JMP 7.01 software (SAS Institute, Cary, NC). Values are expressed as the mean ± SD. Comparisons between two groups were performed by Student's t test. Messenger RNA (mRNA) were split into high- and low-level groups using recursive descent partition analysis of all patients (n = 70) or the GEM group (n = 40), as described by Hoffmann et al. [32]. Categorical variables were compared using the χ2 test (Fisher exact probability test). Survival curves were constructed using the Kaplan-Meier product-limit method and were compared by the log-rank test. To evaluate independent predictive or prognostic factors associated with survival time, multivariate Cox proportional hazards regression analysis was used. Statistical significance was defined as P < .05.

Results

Altered Expression of Genes Encoding Proteins Associated with Gemcitabine Sensitivity in Gemcitabine-Resistant Pancreatic Cancer Cell Lines

Gemcitabine-resistant SUIT-2 (SUIT-2-GR) and Capan-1 (Capan-1-GR) cells were generated by exposure to gradually increasing concentrations of gemcitabine. The IC50 values for gemcitabine of the gemcitabine-resistant cells were significantly higher than those of parental cells, respectively (Table 2). We also quantified the expression levels of genes involved in gemcitabine uptake and metabolism (Figure 1, A–E). SUIT-2-GR cells expressed significantly lower levels of hENT1 and dCK and significantly higher levels of RRM1 and CDA compared with parental cells. Capan-1-GR cells expressed significantly lower levels of dCK and significantly higher levels of CDA compared with parental cells, although the expression levels of CDA were lower than those of SUIT-2 cells. Therefore, alterations of hENT1, dCK, RRM1, and CDA expression were associated with the development of gemcitabine resistance in SUIT-2 cells, whereas only two genes, dCK and CDA, were associated in Capan-1 cells. These data suggest that there are different patterns of gene expression that can develop gemcitabine resistance, and evaluation of several genes is needed to predict gemcitabine sensitivity.

Figure 1
Quantitative analyses of mRNA associated with cellular uptake and metabolism of gemcitabine in gemcitabine-resistant cells. Quantitative analyses of hENT1 (A), dCK (B), RRM1 (C), RRM2 (D), and CDA (E) mRNA in gemcitabine-resistant cell lines (SUIT-2-GR ...
Table 2
IC50 of Each Cell Line.

Univariate and Multivariate Analyses of Each mRNA Level Associated with Gemcitabine Sensitivity and Survival Time

To investigate predictive markers of sensitivity to gemcitabine-based AC in PDAC patients, we quantified hENT1, dCK, RRM1, RRM2, and CDA mRNA levels in 70 FFPE tissue samples of resected PDAC using qRT-PCR. The relationship between gemcitabine-based AC and clinicopathologic characteristics or each mRNA level is summarized in Tables W2 and W3. The GEM and non-AC groups were composed of 40 and 30 cases, respectively. There was no significant correlation between gemcitabine-based AC and the clinicopathologic factors or any of the mRNA levels in PDAC patients. Also, there were no significant differences in the mRNA levels of any of the genes between the two groups (Figure W1, A–E).

Initially, we investigated independent markers that indicated poor prognosis in 70 PDAC patients. We obtained two groups for each gene showing high or low expression, respectively, after normalization to β-actin expression, using cutoff values calculated by recursive descent partition analyses [32] of all patients (N = 70; Table W3). In univariate analyses for OS, the conventional prognostic markers, gemcitabine-based AC (Figure 2; P = .0017), pN status (P = .049), histologic grade (P = .0043), residual tumor (P = .0009), and positive vessel invasion (P = .044) reached significance (Table 3), whereas low dCK (P = .019) and high RRM2 (P = .015) levels normalized to β-actin were associated with a shorter OS (Table 4). Multivariate analysis for OS based on the Cox proportional hazard model was performed on all parameters that were found to be significant on univariate analyses (Table 5). Although OS was significantly dependent on history of gemcitabine-based AC (P = .0001), histologic grade (P < .0001), and R factor (P = .0003), the effects of low dCK and high RRM2 levels did not reach statistical significance. Also, multivariate analysis for DFS (Table W4) showed that DFS was significantly dependent on the histologic grade (P = .0033) and R factor (P < .0001).

Figure 2
Correlation between gemcitabine-based AC and survival time. The patients who received gemcitabine-based AC (GEMgroup) showed a significantly prolongedOStime compared with the non-AC group (P = .0017). *P < .05.
Table 3
Univariate Survival Analysis of Conventional Prognostic Factors (N = 70).
Table 4
Univariate Survival Analysis of mRNA Expression Levels (N = 70).
Table 5
Multivariate OS Analysis (Cox Regression Model) of Conventional Prognostic Factors and mRNA Expression Levels (N = 70).

Next, to determine which parameters are predictive for gemcitabine sensitivity, we evaluated the correlation between each parameter, including gene expression, and DFS in the GEM and non-AC groups. Univariate survival analyses of the GEM group showed that pN status (P = .0052), UICC stage (P = .0066), residual tumor (P = .0002), and positive vessel invasion (P = .018) reached statistical significance for DFS, whereas low dCK (P = .0067) and high RRM2 (P = .003) levels normalized to β-actin were associated with a shorter DFS (Table 6). Low hENT1 (P = .11) and high RRM1 (P = .069) groups tended to associate with a shorter DFS, although these markers did not reach statistical significance (Table 6; Figure 3, A, C, E, and G). In contrast, there was no significant correlation between these gene expression levels and DFS in the non-AC group (Figure 3, B, D, F, and H). Multivariate analysis of the GEM group (Table 7) showed that DFS was significantly dependent on the R factor (P = .0055) and RRM2 level (P = .0055), whereas the effect of low dCK levels did not reach statistical significance.

Figure 3
Correlation between the expression of each mRNA and DFS. Low dCK (P= .0067) and high RRM2 (P=.003) levels, normalized to β-actin, were associated with a shorter DFS in the GEM group (A, C, E, G). In contrast, there was no significant correlation ...
Table 6
Univariate Survival Analysis of Conventional Prognostic Factors and of mRNA Expression Levels in the GEM Group (n = 40).
Table 7
Multivariate DFS Analysis (Cox Regression Model) of Conventional Prognostic Factors and mRNA Expression Levels in the GEM Group (n = 40).

Similarly, we also evaluated the correlation between each parameter and OS in the GEM and non-AC groups. Univariate survival analyses of the GEM group showed that the conventional prognostic markers pN status (P = .014) and residual tumor (P = .0012) reached statistical significance for OS (Table 6). Low hENT1 (P = .011) and dCK (P = .0095) and high RRM1 (P = .041) and RRM2 (P = .030) levels, normalized to β-actin, were associated with a shorter OS (Table 6; Figure 4, A, C, E, and G). In contrast, there was no significant correlation between these gene expression levels and OS in the non-AC group (Figure 4, B, D, F, and H). Multivariate analysis of the GEM group (Table 8) showed thatOS was significantly dependent on pNstatus (P = .029) and R factor (P = .0027), whereas altered gene expression did not reach statistical significance for any gene.

Figure 4
Correlation between the expression of each mRNA and OS. Low hENT1 (P = .011), low dCK (P = .0095), high RRM1 (P = .041), and high RRM2 (P = .030) levels, normalized to β-actin, were associated with a shorterOSin theGEMgroup (A, C, E, G). In contrast, ...
Table 8
Multivariate OS Analysis (Cox Regression Model) of Conventional Prognostic Factors and mRNA Expression Levels in the GEM Group (n = 40).

Furthermore, we created a combined score, which was calculated from each gene expression score determined with recursive descent partition analysis of the GEM group, and evaluated the correlation between survival time and this combined score. Each gene expression was scored as follows: low hENT1, low dCK, high RRM1, and high RRM2, 1; high hENT1, high dCK, low RRM1, and low RRM2, 2. A combined score was created: GEM score = hENT1 score x dCK score x RRM1 score x RRM2 score. As a result, high GEMscore was correlated well with prolonged DFS (Figure 5A) and OS (Figure 5B) in GEM group patients. In univariate analyses of the GEM group, a low GEMscore (<2) was associated with both of a shorter DFS (P = .0003) and a shorter OS (P < .0001). In multivariate analyses of the GEM group, DFS (Table 9) was significantly dependent on UICC stage III/IV (P = .048), R factor (P = .030), and low GEM score (<2, P = .0081), and OS (Table 10) was significantly dependent on pN status (P = .0044), R factor (P = .011), and low GEM score (<2, P = .0002). A low GEM score was an independent predictive and prognostic factor for poor survival in PDAC patients receiving gemcitabine-based AC, with a relative risk of 3.515 and 5.677, respectively.

Figure 5
Correlation between GEM score and survival time. DFS time (A) and OS time (B) after resection of PDAC categorized by combined GEM score (hENT1 score x dCK score x RRM1 score x RRM2 score) in GEM group patients. High GEM scores were well correlated with ...
Table 9
Multivariate DFS Analysis (Cox Regression Model) of Conventional Prognostic Factors and GEM Score in the GEM Group (n = 40).
Table 10
Multivariate OS Analysis (Cox Regression Model) of Conventional Prognostic Factors and GEM Score in the GEM Group (n = 40).

Quantitative Analyses of hENT1, dCK, and RRM2 Expression in Cells Microdissected from Cytologic Specimens

To apply the prediction of gemcitabine sensitivity, based on gene expression levels, to a clinical setting, we quantified hENT1, dCK, RRM1, and RRM2 mRNA levels in cytologic specimens obtained from 15 patients with PDAC who underwent EUS-FNA cytologic examination in our institute. Although a few samples contained abundant neoplastic cells, most samples contained a large amount of blood and inflammatory cells and contained scarce clusters of neoplastic cells (Figure 6, A and B). Therefore, we quantified mRNA levels from WCPs and from microdissected neoplastic cells (laser capture microdissection) prepared from these samples and then compared expression levels between the two preparations. We were unable to detect clear differences in mRNA levels among the WCP samples; however, we could distinguish higher and lower expression levels of each gene among microdissected neoplastic cell samples (Figure 6, C–F). These data suggest that quantification of individual gene expression levels in microdissected neoplastic cells is a potent tool to predict gemcitabine sensitivity even when specimens contain abundant contaminating cells.

Figure 6
Quantitative analyses of mRNA associated with gemcitabine sensitivity in EUS-FNA cytologic specimens. Representative micrographs of cytologic specimens obtained from patients with PDAC who underwent EUS-FNA cytologic examination (A, B). Most samples consisted ...

Discussion

PDAC remains a major therapeutic challenge. Recent randomized clinical trials showed a significant clinical benefit of gemcitabine-based chemotherapy in patients with both resected and unresectable PDAC [7,33]. Therefore, gemcitabine-based chemotherapy remains the standard palliative chemotherapy for PDAC. However, there remains a substantial subset of cases in which gemcitabine-based chemotherapy is insufficient, suggesting the importance of introducing individualized chemotherapy into the clinical setting. Individualized chemotherapy, based on the expression of genes involved in cellular uptake and metabolism of gemcitabine, will be a potent strategy.

We and other investigators have demonstrated that several altered gene expression profiles, including those of hENT1, dCK, RRM1, and RRM2 correlated with the sensitivity to gemcitabine in cancer cell lines [12,15,20,22]. However, analysis of gene expression in two gemcitabine-resistant pancreatic cancer cell lines revealed that there were differences in gene expression between these cancer cell lines. These data suggest that there are different patterns of gene expression that can develop gemcitabine resistance, and combined evaluation of several genes may be required to predict gemcitabine sensitivity.

In the current study, univariate analyses showed that low hENT1, low dCK, high RRM1, and high RRM2 correlated well with poor outcome in patients treated with gemcitabine-based AC, although these altered expression levels did not reach statistical significance in multivariate analysis. Recent clinical studies, including two prospective clinical trials, revealed that PDAC patients with high hENT1 immunoreactivity or high hENT1 expression gained significant benefit from gemcitabine-based AC [13,34,35], and these data are consistent with our results. However, although Akita et al. [36] and Zheng et al. [37] revealed that high RRM1 and high excision repair cross-complementation group 1-expressing patients with PDAC or NSCLC had prolonged survival regardless of AC, Nakahira et al. [20] and Akita et al. [36] also demonstrated that only patients with low RRM1 derive significant benefit from gemcitabine on disease recurrence. Therefore, RRM1 expression may contribute to gemcitabine resistance in PDAC. Moreover, although Sebastiani et al. [38] demonstrated that PDAC patients with high dCK expression had prolonged survival regardless of AC and concluded that genetic alterations of dCK are not a common mechanism of resistance to gemcitabine, previous in vitro studies [15,16,22] support our results showing that high dCK and low RRM2 expressions are correlated with prolonged survival time in PDAC patients who received gemcitabine-based AC. Therefore, to introduce individualized AC into the clinical setting, based on gene expression profiles, the expression levels of several genes will need to be determined, and combined evaluation of these results may be needed. For this reason, we evaluated a simplified score, the GEM score, and found that a low GEM score was predictive for reduced DFS and prognostic for reduced survival in resected PDAC treated with gemcitabine-based AC. However, to evaluate the usefulness of this score, further studies, incorporating larger patient numbers, are required.

In contrast, we found that there was no evident correlation between CDA expression levels and survival time. Recently, a single-nucleotide polymorphism in the CDA gene, which was analyzed using the peripheral blood of cancer patients, was reported to influence the pharmacokinetics and toxicity of gemcitabine [39]. Bengala et al. [40] also demonstrated that high CDA expression and CDA activity levels in peripheral blood mononuclear cells were correlated with shorter survival in gemcitabine-treated patients with advanced pancreatic carcinoma. These data suggest that simple quantification of CDA mRNA in PDAC tissues is not helpful in predicting sensitivity of gemcitabine treatment.

Only 10% to 20% of patients with PDAC are candidates for curative resection [3]; therefore, the remaining 80% to 90% of patients with unresectable advanced PDAC need cytopathologic assessment of EUS-FNA specimens, or pancreatic juice specimens, to predict their sensitivity to chemotherapeutic agents for individualized chemotherapy. The present analysis of mRNA is quantitative (even considering the small amount of specimen available, including cytologic specimens). In addition, the present results revealed that quantification of mRNA in neoplastic cells microdissected from cytologic samples was more useful to distinguish between samples with higher and lower gene expression levels compared with the analysis of WCP samples. The reliability of tests based on tissue or cell extracts is often crucially dependent on the relative abundance of the target cell population, and sampling errors or a large number of “contaminating cells” can lead to false-negative results [26]. hENT1 and dCK mRNA were reported to be expressed in human T lymphocytes and monocytes [41,42], and RRM1 and RRM2 are essential for DNA synthesis in somatic cells. For these reasons, quantification of gene expression in cells microdissected from EUS-FNA cytologic specimens is likely to be useful for predicting gemcitabine sensitivity in patients with PDAC, although further investigations are needed before this approach can be introduced into the clinical setting.

In conclusion, we demonstrated that quantitative analysis of hENT1, dCK, RRM1, and RRM2 mRNA using FFPE tissue samples and evaluation of a combined GEM score were useful in predicting the sensitivity to gemcitabine-based AC in patients with PDAC. In addition, quantitative analysis of these genes in neoplastic cells microdissected from EUS-FNA specimens was useful in determining the treatment for patients with PDAC even when the tumor is unresectable. Quantitative analyses of genes related to cellular uptake and metabolism of cytotoxic agents can be a potent tool to perform individualized chemotherapy.

Supplementary Material

Supplementary Figures and Tables:

Acknowledgments

The authors thank F. Ohkubo (Department of Pathology, Kyushu University Hospital) for skillful cytologic examination and analysis. The authors also thank S. Sadatomi, M. Sato, M. Ohmori, and E. Manabe (Department of Surgery and Oncology, KyushuUniversityHospital) for skillful examination and analysis.

Abbreviations

5′-NT
5′-nucleotidase
AC
adjuvant chemotherapy
CDA
cytidine deaminase
dCK
deoxycytidine kinase
DFS
disease-free survival
EUS-FNA
endoscopic ultrasound-guided fine needle aspiration
FFPE
formalin-fixed paraffin-embedded
hENT1
human equilibrative nucleoside transporter 1
IC50
50% inhibitory concentration
NSCLC
non-small cell lung cancer
OS
overall survival
PDAC
pancreatic ductal adenocarcinoma
qRT-PCR
quantitative real-time reverse transcription-polymerase chain reaction
RR
ribonucleotide reductase
UICC
Union Internationale Contre le Cancer and the American Joint Committee on Cancer
WCP
whole cell pellet

Footnotes

1This work was supported, in part, by the following grants: a grant (H20-Nanchi-Ippan-027) from the Ministry of Health, Labour and Welfare, Japan; a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; and grants from the Japanese Society of Gastroenterology and the Pancreas Research Foundation of Japan. H.F. is a Scientific Research Fellow of the Japan Health Sciences Foundation. The authors declare no competing interests.

2This article refers to supplementary materials, which are designated by Tables W1 to W4 and Figure W1 and are available online at www.neoplasia.com.

References

1. Hirata K, Egawa S, Kimura Y, Nobuoka T, Oshima H, Katsuramaki T, Mizuguchi T, Furuhata T. Current status of surgery for pancreatic cancer. Dig Surg. 2007;24:137–147. [PubMed]
2. Jemal A, Siegel R, Ward E, Hao Y, Xu J, Thun MJ. Cancer statistics, 2009. CA Cancer J Clin. 2009;59:225–249. [PubMed]
3. Matsuno S, Egawa S, Fukuyama S, Motoi F, Sunamura M, Isaji S, Imaizumi T, Okada S, Kato H, Suda K, et al. Pancreatic cancer registry in Japan: 20 years of experience. Pancreas. 2004;28:219–230. [PubMed]
4. Carpelan-Holmstrom M, Nordling S, Pukkala E, Sankila R, Luttges J, Kloppel G, Haglund C. Does anyone survive pancreatic ductal adenocarcinoma? A nationwide study re-evaluating the data of the Finnish Cancer Registry. Gut. 2005;54:385–387. [PMC free article] [PubMed]
5. Wagner M, Redaelli C, Lietz M, Seiler CA, Friess H, Buchler MW. Curative resection is the single most important factor determining outcome in patients with pancreatic adenocarcinoma. Br J Surg. 2004;91:586–594. [PubMed]
6. Neoptolemos JP, Stocken DD, Friess H, Bassi C, Dunn JA, Hickey H, Beger H, Fernandez-Cruz L, Dervenis C, Lacaine F, et al. A randomized trial of chemoradiotherapy and chemotherapy after resection of pancreatic cancer. N Engl J Med. 2004;350:1200–1210. [PubMed]
7. Oettle H, Post S, Neuhaus P, Gellert K, Langrehr J, Ridwelski K, Schramm H, Fahlke J, Zuelke C, Burkart C, et al. Adjuvant chemotherapy with gemcitabine vs observation in patients undergoing curative-intent resection of pancreatic cancer: a randomized controlled trial. JAMA. 2007;297:267–277. [PubMed]
8. Danesi R, De Braud F, Fogli S, Di Paolo A, Del Tacca M. Pharmacogenetic determinants of anti-cancer drug activity and toxicity. Trends Pharmacol Sci. 2001;22:420–426. [PubMed]
9. Sandler AB, Nemunaitis J, Denham C, von Pawel J, Cormier Y, Gatzemeier U, Mattson K, Manegold C, Palmer MC, Gregor A, et al. Phase III trial of gemcitabine plus cisplatin versus cisplatin alone in patients with locally advanced or metastatic non-small-cell lung cancer. J Clin Oncol. 2000;18:122–130. [PubMed]
10. Fukunaga AK, Marsh S, Murry DJ, Hurley TD, McLeod HL. Identification and analysis of single-nucleotide polymorphisms in the gemcitabine pharmacologic pathway. Pharmacogenomics J. 2004;4:307–314. [PubMed]
11. Garcia-Manteiga J, Molina-Arcas M, Casado FJ, Mazo A, Pastor-Anglada M. Nucleoside transporter profiles in human pancreatic cancer cells: role of hcnt1 in 2′,2′-difluorodeoxycytidine-induced cytotoxicity. Clin Cancer Res. 2003;9:5000–5008. [PubMed]
12. Mackey JR, Mani RS, Selner M, Mowles D, Young JD, Belt JA, Crawford CR, Cass CE. Functional nucleoside transporters are required for gemcitabine influx and manifestation of toxicity in cancer cell lines. Cancer Res. 1998;58:4349–4357. [PubMed]
13. Giovannetti E, Del Tacca M, Mey V, Funel N, Nannizzi S, Ricci S, Orlandini C, Boggi U, Campani D, Del Chiaro M, et al. Transcription analysis of human equilibrative nucleoside transporter-1 predicts survival in pancreas cancer patients treated with gemcitabine. Cancer Res. 2006;66:3928–3935. [PubMed]
14. Spratlin J, Sangha R, Glubrecht D, Dabbagh L, Young JD, Dumontet C, Cass C, Lai R, Mackey JR. The absence of human equilibrative nucleoside transporter 1 is associated with reduced survival in patients with gemcitabine-treated pancreas adenocarcinoma. Clin Cancer Res. 2004;10:6956–6961. [PubMed]
15. Ohhashi S, Ohuchida K, Mizumoto K, Fujita H, Egami T, Yu J, Toma H, Sadatomi S, Nagai E, Tanaka M. Down-regulation of deoxycytidine kinase enhances acquired resistance to gemcitabine in pancreatic cancer. Anticancer Res. 2008;28:2205–2212. [PubMed]
16. Blackstock AW, Lightfoot H, Case LD, Tepper JE, Mukherji SK, Mitchell BS, Swarts SG, Hess SM. Tumor uptake and elimination of 2′,2′- difluoro-2′-deoxycytidine (gemcitabine) after deoxycytidine kinase gene transfer: correlation with in vivo tumor response. Clin Cancer Res. 2001;7:3263–3268. [PubMed]
17. Eda H, Ura M, F-Ouchi K, Tanaka Y, Miwa M, Ishitsuka H. The antiproliferative activity of DMDC is modulated by inhibition of cytidine deaminase. Cancer Res. 1998;58:1165–1169. [PubMed]
18. Seve P, Mackey JR, Isaac S, Tredan O, Souquet PJ, Perol M, Cass C, Dumontet C. CN-II expression predicts survival in patients receiving gemcitabine for advanced non-small cell lung cancer. Lung Cancer. 2005;49:363–370. [PubMed]
19. Heinemann V, Xu YZ, Chubb S, Sen A, Hertel LW, Grindey GB, Plunkett W. Cellular elimination of 2′,2′-difluorodeoxycytidine 5′-triphosphate: a mechanism of self-potentiation. Cancer Res. 1992;52:533–539. [PubMed]
20. Nakahira S, Nakamori S, Tsujie M, Takahashi Y, Okami J, Yoshioka S, Yamasaki M, Marubashi S, Takemasa I, Miyamoto A, et al. Involvement of ribonucleotide reductase m1 subunit overexpression in gemcitabine resistance of human pancreatic cancer. Int J Cancer. 2007;120:1355–1363. [PubMed]
21. Rosell R, Felip E, Taron M, Majo J, Mendez P, Sanchez-Ronco M, Queralt C, Sanchez JJ, Maestre J. Gene expression as a predictive marker of outcome in stage IIB-IIIA-IIIB non-small cell lung cancer after induction gemcitabine-based chemotherapy followed by resectional surgery. Clin Cancer Res. 2004;10:4215s–4219s. [PubMed]
22. Duxbury MS, Ito H, Zinner MJ, Ashley SW, Whang EE. RNA interference targeting the m2 subunit of ribonucleotide reductase enhances pancreatic adenocarcinoma chemosensitivity to gemcitabine. Oncogene. 2004;23:1539–1548. [PubMed]
23. Ohuchida K, Mizumoto K, Murakami M, Qian LW, Sato N, Nagai E, Matsumoto K, Nakamura T, Tanaka M. Radiation to stromal fibroblasts increases invasiveness of pancreatic cancer cells through tumor-stromal interactions. Cancer Res. 2004;64:3215–3222. [PubMed]
24. Zhang L, Mizumoto K, Sato N, Ogawa T, Kusumoto M, Niiyama H, Tanaka M. Quantitative determination of apoptotic death in cultured human pancreatic cancer cells by propidium iodide and digitonin. Cancer Lett. 1999;142:129–137. [PubMed]
25. Sobin LH. TNM, sixth edition: new developments in general concepts and rules. Semin Surg Oncol. 2003;21:19–22. [PubMed]
26. Fujita H, Ohuchida K, Mizumoto K, Egami T, Miyasaka Y, Yamaguchi H, Yu J, Cui L, Onimaru M, Takahata S, et al. Quantitative analysis of hTERT mRNA levels in cells microdissected from cytological specimens. Cancer Sci. 2008;99:2244–2251. [PubMed]
27. Nakata K, Ohuchida K, Nagai E, Hayashi A, Miyasaka Y, Kayashima T, Yu J, Aishima S, Oda Y, Mizumoto K, et al. Lmo2 is a novel predictive marker for a better prognosis in pancreatic cancer. Neoplasia. 2009;11:712–719. [PMC free article] [PubMed]
28. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem. 1987;162:156–159. [PubMed]
29. Ohuchida K, Mizumoto K, Ogura Y, Ishikawa N, Nagai E, Yamaguchi K, Tanaka M. Quantitative assessment of telomerase activity and human telomerase reverse transcriptase messenger RNA levels in pancreatic juice samples for the diagnosis of pancreatic cancer. Clin Cancer Res. 2005;11:2285–2292. [PubMed]
30. Abrahamsen HN, Steiniche T, Nexo E, Hamilton-Dutoit SJ, Sorensen BS. Towards quantitative mRNA analysis in paraffin-embedded tissues using real-time reverse transcriptase-polymerase chain reaction: a methodological study on lymph nodes from melanoma patients. J Mol Diagn. 2003;5:34–41. [PubMed]
31. Antonov J, Goldstein DR, Oberli A, Baltzer A, Pirotta M, Fleischmann A, Altermatt HJ, Jaggi R. Reliable gene expression measurements from degraded RNA by quantitative real-time PCR depend on short amplicons and a proper normalization. Lab Invest. 2005;85:1040–1050. [PubMed]
32. Hoffmann AC, Mori R, Vallbohmer D, Brabender J, Klein E, Drebber U, Baldus SE, Cooc J, Azuma M, Metzger R, et al. High expression of HIF1A is a predictor of clinical outcome in patients with pancreatic ductal adenocarcinomas and correlated to PDGFA, VEGF, and BFGF. Neoplasia. 2008;10:674–679. [PMC free article] [PubMed]
33. Burris HA, III, Moore MJ, Andersen J, Green MR, Rothenberg ML, Modiano MR, Cripps MC, Portenoy RK, Storniolo AM, Tarassoff P, et al. Improvements in survival and clinical benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer: a randomized trial. J Clin Oncol. 1997;15:2403–2413. [PubMed]
34. Farrell JJ, Elsaleh H, Garcia M, Lai R, Ammar A, Regine WF, Abrams R, Benson AB, Macdonald J, Cass CE, et al. Human equilibrative nucleoside transporter 1 levels predict response to gemcitabine in patients with pancreatic cancer. Gastroenterology. 2009;136:187–195. [PubMed]
35. Marechal R, Mackey JR, Lai R, Demetter P, Peeters M, Polus M, Cass CE, Young J, Salmon I, Deviere J, et al. Human equilibrative nucleoside transporter 1 and human concentrative nucleoside transporter 3 predict survival after adjuvant gemcitabine therapy in resected pancreatic adenocarcinoma. Clin Cancer Res. 2009;15:2913–2919. [PubMed]
36. Akita H, Zheng Z, Takeda Y, Kim C, Kittaka N, Kobayashi S, Marubashi S, Takemasa I, Nagano H, Dono K, et al. Significance of RRM1 and ERCC1 expression in resectable pancreatic adenocarcinoma. Oncogene. 2009;28:2903–2909. [PubMed]
37. Zheng Z, Chen T, Li X, Haura E, Sharma A, Bepler G. DNA synthesis and repair genes RRM1 and ERCC1 in lung cancer. N Engl J Med. 2007;356:800–808. [PubMed]
38. Sebastiani V, Ricci F, Rubio-Viqueira B, Kulesza P, Yeo CJ, Hidalgo M, Klein A, Laheru D, Iacobuzio-Donahue CA. Immunohistochemical and genetic evaluation of deoxycytidine kinase in pancreatic cancer: relationship to molecular mechanisms of gemcitabine resistance and survival. Clin Cancer Res. 2006;12:2492–2497. [PubMed]
39. Sugiyama E, Kaniwa N, Kim SR, Kikura-Hanajiri R, Hasegawa R, Maekawa K, Saito Y, Ozawa S, Sawada J, Kamatani N, et al. Pharmacokinetics of gemcitabine in Japanese cancer patients: the impact of a cytidine deaminase polymorphism. J Clin Oncol. 2007;25:32–42. [PubMed]
40. Bengala C, Guarneri V, Giovannetti E, Lencioni M, Fontana E, Mey V, Fontana A, Boggi U, Del Chiaro M, Danesi R, et al. Prolonged fixed dose rate infusion of gemcitabine with autologous haemopoietic support in advanced pancreatic adenocarcinoma. Br J Cancer. 2005;93:35–40. [PMC free article] [PubMed]
41. Gamberale R, Galmarini CM, Fernandez-Calotti P, Jordheim L, Sanchez-Avalos J, Dumontet C, Geffner J, Giordano M. In vitro susceptibility of CD4+ and CD8+ T cell subsets to fludarabine. Biochem Pharmacol. 2003;66:2185–2191. [PubMed]
42. Minuesa G, Purcet S, Erkizia I, Molina-Arcas M, Bofill M, Izquierdo-Useros N, Casado FJ, Clotet B, Pastor-Anglada M, Martinez-Picado J. Expression and functionality of anti-human immunodeficiency virus and anticancer drug uptake transporters in immune cells. J Pharmacol Exp Ther. 2008;324:558–567. [PubMed]

Articles from Neoplasia (New York, N.Y.) are provided here courtesy of Neoplasia Press