Fax: (011) 39 9038616
Squamous cell carcinoma of the lung compared with other histotypes shows higher messenger RNA and protein levels for thymidylate synthase
Article first published online: 5 SEP 2006
Copyright © 2006 American Cancer Society
Volume 107, Issue 7, pages 1589–1596, 1 October 2006
How to Cite
Ceppi, P., Volante, M., Saviozzi, S., Rapa, I., Novello, S., Cambieri, A., Lo Iacono, M., Cappia, S., Papotti, M. and Scagliotti, G. V. (2006), Squamous cell carcinoma of the lung compared with other histotypes shows higher messenger RNA and protein levels for thymidylate synthase. Cancer, 107: 1589–1596. doi: 10.1002/cncr.22208
- Issue published online: 18 SEP 2006
- Article first published online: 5 SEP 2006
- Manuscript Revised: 18 JUL 2006
- Manuscript Accepted: 18 JUL 2006
- Manuscript Received: 30 MAY 2006
- Eli Lilly (Indianapolis, IN)
- Squamous cell carcinoma;
- thymidylate synthase;
- real-time polymerase chain reaction;
In patients with cancer, one of the main mechanism of resistance to antimetabolite drugs is related to higher levels of thymidylate synthase (TS) activity.
To investigate the association between TS expression and histopathologic data, 56 resection specimens from patients with nonsmall cell lung carcinoma (NSCLC) were collected consecutively. TS messenger RNA (mRNA) was evaluated in tumor specimens by using real-time polymerase chain reaction (PCR) analysis; protein expression was evaluated by using immunohistochemistry (IHC) in formalin-fixed, paraffin-embedded (FFPE) specimens; and the analysis of TS transcriptional regulation activity was performed by using real-time PCR analysis in snap-frozen normal and tumor specimens.
The amplification of the TS gene from FFPE tissues was obtained from all samples, with a median level (unit-less ratio) of 1.45 (range, 0.34–5.24); whereas positive TS status at IHC (>10% positive cells) was detected in 56% of samples. It is noteworthy that TS expression was significantly higher in squamous cell carcinoma compared with adenocarcinoma when both mRNA levels (2.17 vs. 1.16; P < .0001) and protein levels (P = .0269) were considered in FFPE specimens, and a strong association was observed between mRNA and protein expression (P = .00017). Moreover, higher TS levels were observed in high-grade tumors (P = .0389 and P = .0068 for mRNA and protein quantification, respectively). The analysis in snap-frozen samples revealed that the TS gene was up-regulated strongly in tumors (P = 3.8 × 10−12), and an 8-fold increase (as a cut-off value) in the TS mRNA ratio between tumor and corresponding normal tissue was detected in 32 of 56 patients (57%) bearing preferentially squamous cell tumors (P = .0022) and high-grade tumors (P < .001).
Data from the current study consistently indicated higher TS expression levels in squamous cell and in high-grade carcinomas. This information may be useful in selecting which patients with NSCLC should receive treatment with TS-inhibiting agents. Cancer 2006. © 2006 American Cancer Society.
Despite extensive preclinical and clinical research, lung cancer remains the leading cause of cancer-related death in many countries.1 The responsiveness of patients with nonsmall cell lung cancer (NSCLC) to combination chemotherapy is related to clinical variables, such as stage, performance status, and weight loss; and, in patients with metastatic disease, the response rate to combination chemotherapy ranges from 20% to 40%.2, 3
In the clinical setting, the evaluation of messenger RNA (mRNA) expression levels of selected genes potentially may enable clinicians to tailor chemotherapy according to each individual's gene profile and to produce a substantial improvement in the therapeutic outcome in terms of overall survival, time to progression, and response to therapy.4 Thymidylate synthase (TS) is an enzyme that plays an important role in DNA biosynthesis, catalyzing the methylation of fluorodeoxyuridine monophosphate (dUMP) to deoxythimidine monophosphate (dTMP), and is the target enzyme for antimetabolite agents, such as 5-fluorouracil (5-FU). For this reason, TS has been investigated widely and currently is considered among the best known drug targets in the anticancer area.5 Several studies clearly showed the efficacy of the evaluation of intratumoral expression of TS mRNA in predicting the response to 5-FU therapy in patients with breast cancer,6 colorectal cancer,7 head and neck cancer,8 pancreatic cancer,9 and NSCLC.10 This association was not reported invariably11, 12; however, in most of those studies, an adverse effect of higher TS levels on prognosis was documented in different types of human cancers.6–10 In patients with colorectal cancer, it was reported that low microsatellite instability, the genotype 2R/2R of the 5′-untranslated region-enhancer region of the TS gene, and lower intratumoral expression levels were associated with a good response to chemotherapy.13
Recently, pemetrexed, which is a multitargeted antifolate, showed cytotoxic activity in different types of human cancers.4 This agent reportedly is a potent TS and, although less intensely, a dihydrofolate reductase and glycinamide ribonucleotide formyltransferase inhibitor.14 Pemetrexed already has been approved for the treatment of malignant mesothelioma15 and second-line NSCLC.16
The quantification of TS mRNA levels reportedly was useful in determining the prognosis of patients with NSCLC and in the immunohistochemical (IHC) evaluation of TS protein expression, which predicted survival well in untreated patients.17TS protein expression was correlated significantly with higher proliferative activity of NSCLC cells and, consequently, with a poor prognosis in patients with NSCLC who had higher TS level.18 No correlations between intratumoral TS levels and any known clinicopathologic variables were reported with the exclusion of a recently published article in which TS gene expression was associated with disease stage, lymph node metastasis, tumor differentiation, prognosis, and tumor cell proliferation in patients with lung adenocarcinoma.19
The current study was designed to verify the potential of different techniques for the detection of TS gene and protein expression in patients with chemotherapy-naive NSCLC, to define the differential expression of TS in normal and neoplastic lung tissue, and, finally, to verify whether TS expression is correlated with any of the known clinicopathologic features of NSCLC.
MATERIALS AND METHODS
Patients and Samples
Fifty-six surgical samples were collected from consecutive patients with resectable NSCLC (American Joint Committee on Cancer Stages I-IIIA) who were treated at the University of Torino, San Luigi Hospital, Orbassano (Turin, Italy) from January 2003 to February 2004. The study was approved by the hospital's institutional review board. All samples were rendered anonymous by a pathologist who was not taking part in the study. None of the researchers who conducted the gene and protein expression analyses in the study had access to clinicopathologic data. Informed consent was obtained from all patients. All analyses were conducted on the surgically resected samples from which the pathologists selected a representative area of the tumor. One part of the tumor specimen and the corresponding normal lung obtained from the same lobe or lung (as judged by macroscopic visual assessment) was snap frozen in liquid nitrogen, and the immediately adjacent part was fixed in formalin, embedded in paraffin, and used for gene expression and immunohistochemical analyses.
Microdissection, RNA Isolation, and Combinational DNA Synthesis in Paraffin-Embedded Tissues
From each paraffin block of representative tumor area, serial sections with a thickness of 10 μm were prepared and then stained with nuclear Fast Red (Sigma-Aldrich, St. Louis, MO). Malignant cells were selected under microscopic magnification (from × 5 to × 10) and were dissected from the slide simply by using a scalpel.
RNA isolation and combinational DNA synthesis
RNA isolation was performed according to a proprietary procedure (U.S. Patent No. 6,248,535). In brief, tissue samples were heated at 92°C for 30 minutes in 4 mol/L dithiothrietol-guanidinium isothiocyanate/ sarcosine (4 mol/L guanidinium isothiocyanate; 50 mmol/L Tris-HCl, pH 7.5; 25 mmol/L ethylenediamine tetraacetic acid [EDTA]; Invitrogen, Carlsbad, CA). Fifty microliters of 2 mol/L sodium acetate, pH 4.0, followed by 600 μL of freshly prepared phenol/chloroform/isoamyl alcohol (250:50:1 dilution) were added to the tissue suspension. The suspension was centrifuged at × 13,000 rounds per minute (rpm) for 8 minutes in a chilled (8°C) centrifuge. The upper aqueous phase was removed and combined with glycogen (10 μL) and 300 to 400 μL of isopropanol. The tubes were placed at − 20°C for 30 to 45 minutes to precipitate the RNA. After centrifugation at 13,000 rpm for 7 minutes in a chilled (8°C) centrifuge, the supernatant fluid was carefully poured off, the pellet was resuspended in 50 μL of 5 mmol/L Tris, and combinational DNA (cDNA) synthesis was performed as described previously.20
RNA extraction and cDNA synthesis in snap-frozen specimens
Total RNA (totRNA) was isolated from lung specimens using the RNeasy 96 Kit (Qiagen, Hilden, Germany) implemented on Biorobot 8000 (Qiagen, Hilden, Germany) according to the manufacturer's instructions. RNA was extracted from tumor specimens (15–25 mg) and from normal lung tissue specimens (60–80 mg). Genomic DNA contamination was removed by using an insolubilized-DNAseI treatment. TotRNA then was quantified by using a spectofotometer, and 2 μg of totRNA were retrotranscribed with random hexamer primers by using the cDNA Archive Kit (Perkin Elmer, Boston, MA) according to the manufacturer's suggestions.
Real-Time Polymerase Chain Reaction Analysisof Snap-Frozen and Formalin-Fixed, Paraffin-Embedded Specimens
Expression levels of TS and 2 reference genes, polymerase RNA II polypeptide A (POLR2A) and 18S ribosomal RNA (rRNA 18S), were evaluated using commercially available TaqMan Probes (“Assay on Demand”; Applied Biosystems, Foster City, CA) with optimized primer and probe concentrations (TS, Hs00426591_m1; POLR2A, Hs00172187_m1; rRNA18S, Hs99999901_s1). The selection of POL2RA and rRNA 18S as housekeeping genes was related to the results of a preliminary validation assessment of these 2 genes that indicated a lower level of intrasample and intersample variability between normal and neoplastic lung tissues (unpublished data). Quantitative polymerase chain reaction (qPCR) was performed on an ABI PRISM 7900HT Sequence Detection System (Applied Biosystems) in 384-well plates assembled by Biorobot 8000 (Qiagen), and the reaction was performed in a final volume of 20 μL. All qPCR mixtures contained 1 μL of cDNA template (corresponding to approximately 20 ng retrotranscribed totRNA), 1 × TaqMan Universal PCR Master Mix (× 2; Applied Biosystems), and 1 × Assay-on-Demand Gene Expression Assay Mix (× 20). Cycle conditions were as follows: after an initial hold of 2 minutes at 50°C to allow AmpErase-UNG activity and 10 minutes at 95°C, the samples were cycled 40 times at 95°C for 15 seconds and at 60°C for 1 minute. Baseline and threshold for cycle threshold (Ct) calculation were set manually with ABI Prism SDS 2.1 software.
The fold change in TS expression levels between malignant and nonmalignant paired samples were evaluated as 2-ΔΔCt 21 normalized to the average of the ΔΔCt (AvΔΔCt) values obtained for POL2RA and rRNA 18S. Genes were considered to be expressed differentially if the final AvΔΔCt was >1 (i.e., a 2-fold change).
Relative cDNA quantification of TS and of an internal reference gene (β-actin) was done by using a fluorescence-based real-time detection method that was described previously.22, 23 The gene expression analysis in formalin-fixed, paraffin-embedded (FFPE) tissue was performed using a different set of primers because of the lower amplicon size requested to overcome the limits of RNA degradation. For the same reason, a different internal reference gene (β-actin) was used. In addition, the current study was not designed to make a direct comparison between gene quantification in snap-frozen and FFPE tissue. The sequences of the primers and probes used were as follows: TS: forward, 5′-GGCCTCGGTGTGCCTTT-3′; re verse, 5′-GATGTGCGCAATCATGTACGT-3′; probe, (FAM)-5′-AACATCGCCAGCTACGCCCTGC-3′-(TAMRA); β-actin: forward, 5′-TGAGCGCGGCTACAGCTT-3′; reverse, 5′-TCCTTAATGTCACGCACGATTT-3′, probe, (FAM)-5′-ACCACC ACGGCCGAGCGG-3′-(TAMRA).24 The PCR reaction mixture consisted of 1200 nmol/L of each primer; 200 nmol/L probe; 200 nmol/L each of deoxyadenosine triphosphate, deoxycyitidine triphosphate, deoxyguanine triphosphate, and deoxythymidine triphosphate; 3.5 mmol/L MgCl2; and 1 × Taqman Universal PCR Master Mix to a final volume of 20 μL (all reagents were from PE Applied Biosystems). Cycling conditions were 50°C for 2 minutes and 95°C for 10 minutes followed by 46 cycles at 95°C for 15 seconds and 60°C for 1 minute. Relative gene expression levels are expressed as unitless ratios between 2 absolute measurements (genes of interest/internal reference gene). Colon, liver, and total RNAs (all from Stratagene, La Jolla, CA) were used as control calibrators on each plate.
Immunohistochemistry in formalin-fixed, paraffin-embedded specimens
From each paraffin block, 5-μm-thick sections were prepared and stained with hematoxylin and eosin for conventional histologic examination. In addition, serial sections collected on charged slides were analyzed by immunohistochemistry (IHC) staining, as follows: After deparaffinization and rehydration through graded alcohols and phosphate-buffered saline (pH 7.5), endogenous peroxidase activity was blocked by absolute methanol and 0.3% hydrogen peroxide for 15 minutes. The slides were incubated for 40 minutes at room temperature with primary mouse anti-TS antibody (dilution, 1:100; Zymed, South San Francisco, CA). The immune reaction was revealed in a biotin-free detection system based on a dextran chain linked to the secondary antibody and that carried peroxidase (En Vision; Dako, Glostrup, Denmark), using 3,3′-diaminobenzidine (Dako) as the chromogen. For antigen retrieval steps, the sections were treated in a microwave oven for 15 minutes in EDTA buffer (pH 8.0). Slides were counterstained with hematoxylin, dehydrated, and mounted, as described previously.25 The immune reaction was scored as negative (totally negative, score 0) or positive, in a semiquantitative, 3-tie system based on the extent of reactivity (score 1, 1–10% reactivity; score 2, 11–50% reactivity; and score 3, >50% reactivity).
The Mann–Whitney U test was used to test significant associations between the continuous variable gene expression and dichotomous variables (patient age, gender, etc.) and to test the correlation between TS mRNA levels in FFPE tissue sections and IHC analyses. The Kruskal–Wallis test was used to test significant associations between gene expression and multiple variables, such as histology, tumor stage, tumor grade, pathologic tumor classification (pT), pathologic lymph node status (pN), and IHC staining score. The chi-square test was used to test the significance of the IHC results. Wilcoxon and chi-square tests were used to determine the significance of the results of the evaluation of transcriptional regulation activity. Statistical significance was set at P = .05.
Tumor and Patient Characteristics
The median age of the 56 patients who were included in the current study was 70.5 years (range, 43–84 years), there were 42 males and 14 females, and the extent of disease ranged from Stage IA to Stage IIIA. Histologically, 30 patients had a diagnosis of adenocarcinoma, 21 patients had a diagnosis of squamous cell carcinoma, 3 patients of large cell carcinoma, and 2 patients of bronchioloalveolar carcinoma. Histopathologic grades (Grades 1, 2 and 3) indicated well differentiated, moderately differentiated, and poorly differentiated tumors, respectively. All patient characteristics are shown in Table 1.
|Characteristic||No. of patients||Percentage||Median TS level*||P|
|Age (median, 70.5 y; range, 43–84 y)|
|Pathologic tumor classification|
|Pathologic lymph node status|
TS mRNA Expression Is Higher in Squamous Cell Carcinoma and in High-Grade NSCLC
The quantification of mRNA from FFPE tumor specimens was performed in real-time PCR, and the results were compared by using β-actin as the internal reference gene. TS expression levels (in unit less ratio) ranged from 0.34 to 5.24 (median, 1.45; mean, 1.72; standard deviation 1.24), and a significant correlation was observed between histology and TS mRNA levels. In fact, the median expression level for adenocarcinoma was 1.16 whereas for squamous cell carcinoma the median expression level was 2.17 (P < .0001) (Fig. 1). We also documented higher TS expression levels in high-grade tumors (Grade 3, poorly differentiated; P = .0389). Histopathologic Grade 3 included 8 adenocarcinomas, 10 squamous cell carcinomas, and 3 large cell carcinomas. No other significant associations were observed with variables like age (P = .209; adopting the median as a cut-off value), gender (P = .604), tumor stage (P = .503), pT (P = .146), or pN (P = .697) and TS mRNA expression. The results are shown in Table 1.
Higher TS Protein Expression in Squamous Cell Carcinoma and High-Grade NSCLC
Comparable results (Table 2) were documented in the IHC analyses. Ten patients (18%) had a negative score (0), 15 patients (27%) had a score of 1, 21 patients (37%) had a score of 2, and 10 patients (18%) had a score of 3. Patients were grouped as negative (scores of 0 and 1) or positive (scores of 2 and 3) by using 10% positive cells as a cut-off value. Consequently, 31 patients (55%) were TS positive, and 25 patients (45%) were TS negative. Patients who had squamous cell carcinoma had higher TS expression compared with patients who had adenocarcinoma (P = .0269), and patients who had Grade 3 tumors had higher TS expression compared with patients who had Grade 1 and 2 tumors (P = .0068). No significant associations were observed between TS IHC expression levels and pT or pN status. Figure 2 shows IHC staining with the anti-TS antibody in 2 representative sections of adenocarcinoma and squamous cell carcinoma.
|Characteristic||TS status (No. of patients)*||P|
|Pathologic tumor classification|
|Pathologic lymph node status|
Correlation between TS mRNA Levels and Protein Expression in Formalin-Fixed, Paraffin-Embedded Specimens
A strong correlation was observed between TS mRNA and protein levels: the median gene expression was 0.95 for the group of TS-negative and 1.94 for the group of TS-positive tumors at IHC (P = .00017; Mann–Whitney U test) (Fig. 3). The median mRNA level for each IHC score was 0.986, 0.907, 1.857, and 1.970, respectively, for a score of 0, 1, 2, and 3 (P = .00642; Kruskal–Wallis test).
In Snap-Frozen Specimens TS mRNA Is Up-Regulated in Tumor Compared with Normal Lung Especially in Squamous Cell NSCLC
The analysis of TS transcript regulation in snap-frozen specimens by qPCR clearly showed that TS expression levels in tumor were considerably higher compared with the levels in normal tissues (P = 3.8 × 10− 12; Wilcoxon test) (Fig. 4). In fact, in 52 of 56 patients (92.3%), TS mRNA was up-regulated at least 2-fold in tumor tissue related to normal lung (results not shown). Because of the general over-expression of TS transcript in tumor tissues, we adopted the more stringent cut-off level of an 8-fold change, and a high TS up-regulation (i.e., >8-fold) was observed in 32 of 56 patients (57.1%). These patients were affected preferentially by squamous cell carcinoma (P = .0022) and high-grade cancers (P < .001; chi-square test). All data are shown in Table 3.
|Characteristic||TS up-regulation (No. of patients)||Total||P|
|High (FC ≥≥ 8)||Low (FC << 8)|
|Total no. of patients||32||24||56|
|Pathologic tumor classification|
|Pathologic lymph node status|
The data from the current study indicate higher TS expression levels in squamous cell and in high-grade carcinomas. This differs from previously published results from other investigators,17, 18 who observed that any association between TS mRNA levels and/or IHC positivity and histopathologic features did not reach the level of statistical significance. By contrast, in the current study, we observed higher expression of TS in squamous cell carcinoma compared with to nonsquamous histotypes. This was determined independently by using 2 different techniques: the quantification of mRNA by real-time PCR and the evaluation of protein levels in FFPE neoplastic tissues, although there was a statistically significant difference between the 2 methods (P < .0001 vs. P < .0269, respectively). Moreover, in the current study, a strong correlation was observed between TS transcript and protein expression (P = .00017). In parallel to expression data on formalin-fixed, paraffin-embedded tumor samples, in the transcriptional regulation analysis of TS gene in snap-frozen samples, we were able to identify those tumors that had high or low transcriptional TS modulation activity. The general up-regulation of TS in the tumor tissues rather than normal tissues reported here was not surprising and probably was because of the higher proliferation rate of tumor cells compared with normal cells. Similar to what was demonstrated in the analysis of FFPE specimens by using a different internal reference gene, higher transcriptional regulation was observed in squamous cell carcinoma.
Furthermore, the evidence of higher TS mRNA levels in high-grade tumors was confirmed by IHC. This result agrees with what was reported previously in a series of patients with adenocarcinoma only19 and probably is associated with the proliferation and differentiation rate in NSCLC, thus suggesting a possible role for TS as a marker of tumor aggressiveness.
Combination chemotherapy with cisplatin and 5-FU (a TS-specific inhibitor drug) have been rarely investigated in patients advanced NSCLC, but scant data from a relatively large Phase II study indicate a comparable level of activity but lower hematologic toxicity than what usually are observed with currently used cisplatin-based regimens.26, 27 It is noteworthy that adjuvant chemotherapy with uracil-tegafur (UFT), an oral 5-FU derivative, reportedly improved survival significantly among Japanese patients with completely resected NSCLC, as reported in several trials and in a meta-analysis.28, 29 Moreover, in a Japanese trial, adjuvant UFT was effective only in patients with early-stage adenocarcinoma, and no significant benefit to survival was reported in patients with squamous cell carcinoma.30 In all of the studies mentioned above, TS activity was not assessed; however, the results of the current study provide a rational explanation for the better outcome and response to chemotherapy among patients with adenocarcinoma of the lung who received TS-inhibitor regimens.
A new multitargeted antifolate, pemetrexed, is used increasingly in the management of malignant pleural mesothelioma and second-line NSCLC, and the evaluation of its activity in combination with cisplatin as front-line treatment of metastatic NSCLC currently is underway. Ongoing Phase II studies are evaluating the activity of this drug in bladder, head and neck, esophageal, renal, and cervical carcinomas.31 Although the Phase III study in NSCLC comparing second-line pemetrexed with docetaxel did not reveal a higher level of activity of pemetrexed in specific subgroups of patients,16 the high prescription cost of the drug itself suggests the necessity to identify which patients are more likely to benefit from the treatment. Translational research is underway with the objective of characterizing patients and tumors on a molecular basis; and, by identifying which patients may benefit from a certain therapy. Pharmacogenomic diagnostic techniques may be able to predict patient outcomes accurately and to tailor treatment strategies according to tumor genotype and phenotype.32 Several investigators worldwide have investigated extensively and reported on the usefulness of the intratumoral TS assessment in predicting the response to 5-FU and other TS-inhibiting drugs. In the current study, we tested some of these new diagnostic tools, and the results indicated a strong correlation between TS, histology, and histopathologic grade. In conclusion, these data potentially may offer clinicians an easy and reliable tool for the selection of patients who are eligible for TS inhibitor-based therapy for the treatment of NSCLC. Further validation of the experimental hypothesis will come only from specifically designed, prospective studies that address the predictive and prognostic value of TS assessment in patients with NSCLC.