Lack of correlation between immunohistochemical expression of E2F-1, thymidylate synthase expression and clinical response to 5-fluorouracil in advanced colorectal cancer

O. Belvedere1, F. Puglisi1, C. Di Loreto2, P. Cataldi2, A. Guglielmi3, C. Aschele4 and A. Sobrero3,*

1 Division of Medical Oncology, University of Udine, P. le S. M. Misericordia, Udine; 2 Department of Pathology, University of Udine, P. le S. M. Misericordia, Udine; 3 Division of Medical Oncology, S. Martino General Hospital, L. go R. Benzi 10, Genoa; 4 Division of Medical Oncology, Galliera General Hospital, Mura delle Cappuccine 14, Genoa, Italy

Received 25 June 2003; accepted 3 September 2003


    ABSTRACT
 Top
 ABSTRACT
 Introduction
 Materials and methods
 Results
 Discussion
 REFERENCES
 
Background:

The level of the enzyme thymidylate synthase (TS) is known to inversely correlate with the clinical activity of 5-fluorouracil (FU) in advanced colorectal cancer patients. Since the correlation is not very strong, we have retrospectively analyzed the expression of E2F-1 in tumor samples or metastases from 25 patients with advanced colorectal cancer, homogeneously treated with an FU-based regimen. E2F-1 is a transcription factor regulating the expression of TS along with other crucial DNA synthesis related enzymes.

Materials and methods:

E2F-1 expression was analyzed by immunohistochemistry using the anti-E2F-1 monoclonal antibody KH95, scoring 2000 cells/case. Expression of TS was evaluated by immunohistochemistry using a rabbit anti-human polyclonal antibody.

Results:

The level of E2F-1 expression did not correlate with TS expression, although a trend for correlation between E2F-1 level and maximal tumor shrinkage was observed (r = 0.42; P = 0.054).

Conclusions:

In spite of previous reports demonstrating that E2F-1 quantified by rt-PCR and western blot correlates with TS and could be used as a predictor to select colorectal cancer patients more likely to respond to FU treatment, our data suggest that, under these experimental conditions, immunohistochemistry cannot be used for such selection.

Key words: colorectal cancer, E2F-1, immunohistochemistry, thymidylate synthase


    Introduction
 Top
 ABSTRACT
 Introduction
 Materials and methods
 Results
 Discussion
 REFERENCES
 
Despite the demonstrated activity of CPT-11 and oxaliplatin against colorectal cancer, 5-fluorouracil (FU) continues to be the main component of combination chemotherapy in this disease. Based on the correlation between the levels of thymidylate synthase (TS) and the response to treatment with fluoropyrimidine both in vitro and in vivo, the expression of TS has been identified as a predictor for response to 5-FU. Recently, we and others have demonstrated that high levels of TS inversely correlate with the response rate, the time to progression and the overall survival of advanced colorectal cancer patients treated with FU [14]. In our study, immunohistochemical TS quantitation was performed on archival specimens from 48 patients homogeneously treated with bolus FU alternating with continuous infusion FU [3]. We concluded that immunohistochemical assessment of TS provides a convenient and low-cost technique to identify patients unresponsive to TS inhibitors. However, the correlation between TS and response is incomplete, and the extent of FU benefit is limited even in patients expressing low TS.

Several molecules are under investigation as potential predictors for response to fluoropyrimidines. One of these is the transcription factor E2F-1, which controls the transcription of several genes encoding proteins involved in DNA synthesis, namely TS, dihydrofolate reductase, thymidine kinase, and ribonucleotide reductase (Figure 1) [57]. In vivo, a high correlation between E2F-1 and TS was recently shown in colorectal metastases to the lung and the liver, using quantitative rt-PCR [8].



View larger version (22K):
[in this window]
[in a new window]
 
Figure 1. Schematic representation of the pRb/E2F-1 pathway.

 
In this study, we have assessed whether E2F-1 expression determined by immunohistochemistry correlates with TS expression and the clinical response to FU in a search for a better predictor of treatment outcome than TS.


    Materials and methods
 Top
 ABSTRACT
 Introduction
 Materials and methods
 Results
 Discussion
 REFERENCES
 
Patients and sample characteristics
Paraffin sections of primary colorectal tumors or metastases were obtained from 25 patients homogeneously treated with a regimen alternating bolus and continuous infusion FU [3].

TS immunohistochemical analysis
TS expression data were retrieved from previous analyses, performed as reported previously [3]. Briefly, tissue sections were heated in a microwave oven at 300 W for 10 min, cooled and stored in TBS at pH 7.6. To block non-specific binding of the primary antibody, a normal rabbit serum (DAKO X901; Dako, Glostrup, Denmark) dilution in TBS was used for 20 min. Sections were incubated with a rabbit polyclonal antibody to recombinant human TS (2 µg/ml) for 60 min in a humidified chamber at room temperature. Slides were then incubated with biotinylated swine anti-rabbit immunoglobulins for 20 min (DAKO-E353), followed by the avidin–biotinylated peroxidase complex for 30 min. After the color reaction product was developed with diaminobenzidine chromogen solution for 5 min, slides were counterstained with light hematoxylin for 1 min, dehydrated in a series of ethanols, cleared in xylene, and then examined under a light microscope. Only tumor cells with cytoplasmatic staining were counted as positive. TS expression was quantitated using a visual grading system based on the intensity of staining and classified into five groups from negative to very high intensity of staining. When heterogeneous levels of TS expression were found within a tumor (in multiple sections from different paraffin-embedded blocks of the same tumor), the level of TS expression of that lesion was defined according to the highest TS score.

E2F-1 expression
The expression of E2F-1 was retrospectively assessed by immunohistochemistry, following previously reported methods [9]. Briefly, sections were incubated in 0.01 M citrate buffer at pH 6.0 and heated twice in a microwave oven for 5 min per cycle. Staining was performed with the mouse anti-E2F-1 mAb KH95 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), used at 1:100 dilution. Reactivity was detected by an avidin–biotin immunoperoxidase detection system. Tonsil sections were processed in parallel, as a positive control. Counts were performed with a light microscope at 40x magnification, scoring 2000 epithelial cells/case. Only nuclear staining was considered positive, regardless of signal intensity. Tumor cells exhibiting brown reaction product were scored as positive, regardless of the staining intensity.


    Results
 Top
 ABSTRACT
 Introduction
 Materials and methods
 Results
 Discussion
 REFERENCES
 
Patient and sample characteristics are reported in Table 1. Of the 25 samples, 17 were from the primary site and eight were from metastases. Within this highly selected cohort of patients, the response rate was 44%, as assessed by radiological evaluation according to the WHO criteria.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient and sample characteristics
 
There was no statistically significant difference among the percentage of E2F-1-positive cells observed in colorectal cancer samples from different sites, as shown in Figure 2. The mean percentage of E2F-1-positive cells was similar in tissue samples from colonic, rectal primaries and metastatic lymph nodes: 28% for colon (14 patients), 36% for rectal (three patients), and 34% for the lymph nodes (seven patients).



View larger version (14K):
[in this window]
[in a new window]
 
Figure 2. E2F-1-positive cells in different colorectal cancer samples. Mean values are also shown.

 
To determine whether immunohistochemical quantitation of E2F-1 correlates with TS levels, we compared E2F-1 expression to the TS expression we had previously measured on the same samples [3]. No correlation was observed between TS and E2F-1 expression in terms of percentage of positive cells (r = 0.02, P = 0.92) (Figure 3). Analyzing these data by site, there was no correlation either in the primary sites (r = –0.30, P = 0.26) or in the metastases (r = 0.28, P = 0.52). In addition, we also evaluated the correlation of E2F-1 level with maximal tumor shrinkage, measured as a ratio between the initial tumor area and that measured at the time of maximal response. As shown in Figure 4, although not significant, a trend for correlation between E2F-1 level and maximal tumor shrinkage was observed (r = 0.42, P = 0.054).



View larger version (14K):
[in this window]
[in a new window]
 
Figure 3. Correlation between E2F-1 and TS expression broken down by objective response.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 4. Correlation of E2F-1 level with tumor shrinkage (20 patients).

 

    Discussion
 Top
 ABSTRACT
 Introduction
 Materials and methods
 Results
 Discussion
 REFERENCES
 
Since 5-fluorodeoxyuridylate, the active metabolite of FU, needs to bind covalently to TS in the presence of 5,10-methylene-tetrahydrofolic acid, it is biologically plausible that low levels of TS would predict sensitivity to the fluoropyrimidine, while high levels would predict resistance, as demonstrated by several correlation studies carried out in the advanced setting of colorectal cancer. In a search for a more precise prediction of treatment outcome, Salonga et al. have added to TS the determination of two other potential markers of FU sensitivity: dihydropyrimidine dehydrogenase and thymidine phosphorylase [10]. They found that when the tumor sample had low TS, low dihydropyrimidine dehydrogenase and low thymidine phosphorylase expression levels, the response rate to infusional FU was 100%. However, the overall survival and the shape of the survival curves were not significantly different as a function of expression level. This study was based on molecular biology methods. However, for routine evaluation of predictors of response to chemotherapy on a large number of patients, immunohistochemistry presents several advantages compared to molecular biology: it is simple, inexpensive and fast.

In a search for easier methods of prediction of resistance or sensitivity, the immunohistochemical determination of E2F-1 seems particularly appealing.

Consistent with the biochemical cascade that leads to TS expression, a previous study showed that E2F-1 expression correlates with TS expression in colorectal cancer tissues, as determined by quantitative rt-PCR [8]. In addition, Banerjee et al. demonstrated that induced overexpression of E2F-1 in a human fibrosarcoma cell line results in increased levels of TS and resistance to FU [11]. Based on these studies, E2F-1 should be further investigated to confirm TS expression level in the selection of patients sensitive to FU.

The results of our study are negative both in terms of correlation between E2F-1 and TS, and prediction of treatment outcome. Aside from the small number of patients, there are at least two explanations for this potentially false-negative result. E2F-1 determination was done 2 years after the original TS determination and the effect of time could be substantial, as demonstrated for p53 [12]. As an alternative, immunohistochemistry does not discriminate between the acetylated and non-acetylated forms of E2F-1. As recently reported, E2F-1 activity is regulated by acetylation: the acetylated form of E2F-1 has an increased DNA-binding ability, a higher transactivation potential and an increased stability [13, 14]. The level of acetylated E2F-1 might correlate with TS and treatment outcome. Still, to quantitate the acetylated E2F-1, immunoblotting would be required, making this procedure time-consuming and expensive, and therefore not suitable for routine use.

In conclusion, in spite of the previously reported correlation between E2F-1 and TS expression by rt-PCR and western blot, suggesting E2F-1 as a predictor to select colorectal cancer patients more likely to respond to FU, our results suggest that under these experimental conditions immunohistochemistry may not be a reliable technique to assess E2F-1 level for such a selection.


    Acknowledgements
 
This work was partially supported by a grant from CNR Bridge (27 December 1997, no. 449), Progetto Strategico MURST, and two grants from Associazione Italiana per la Ricerca sul Cancro (AIRC 2000 and AIRC 2001).


    FOOTNOTES
 
* Correspondence to: Prof. A. Sobrero, Medical Oncology, S. Martino Hospital, Largo R. Benzi, 10, 16132 Genoa, Italy. Tel: +39-010-555-3977; Fax: +39-010-555-5141; E-mail: alberto.sobrero{at}hsanmartino.liguria.it Back


    REFERENCES
 Top
 ABSTRACT
 Introduction
 Materials and methods
 Results
 Discussion
 REFERENCES
 
1. Johnston PG, Lenz HJ, Leichman CG et al. Thymidylate synthase gene and protein expression correlate and are associated with response to 5-fluorouracil in human colorectal and gastric tumors. Cancer Res 1995; 55: 1407–1412.[Abstract]

2. Leichman CG, Lenz HJ, Leichman L et al. Quantitation of intratumoral thymidylate synthase expression predicts for disseminated colorectal cancer response and resistance to protracted-infusion fluorouracil and weekly leucovorin. J Clin Oncol 1997; 15: 3223–3229.[Abstract]

3. Aschele C, Debernardis D, Casazza S et al. Immunohistochemical quantitation of thymidylate synthase expression in colorectal cancer metastases predicts for clinical outcome to fluorouracil-based chemotherapy. J Clin Oncol 1999; 17: 1760–1770.[Abstract/Free Full Text]

4. Edler D, Hallstrom M, Johnston PG et al. Thymidylate synthase expression: an independent prognostic factor for local recurrence, distant metastasis, disease-free and overall survival in rectal cancer. Clin Cancer Res 2000; 6: 1378–1384.[Abstract/Free Full Text]

5. DeGregori J, Kowalik T, Nevins JR. Cellular targets for activation by the E2F-1 transcription factor include DNA synthesis and G1/S regulatory genes. Mol Cell Biol 1995; 15: 4215–4224.[Abstract]

6. Wu L, Timmers C, Maiti B et al. The E2F1-3 transcription factors are essential for cellular proliferation. Nature 2001; 414: 457–462.[CrossRef][ISI][Medline]

7. Brehm A, Miska E, Reid J et al. The cell cycle-regulating transcription factors E2F-RB. Br J Cancer 1999; 80 (Suppl 1): 38–41.[CrossRef]

8. Banerjee D, Gorlick R, Liefshitz A et al. Levels of E2F-1 expression are higher in lung metastasis of colon cancer as compared with hepatic metastasis and correlate with levels of thymidylate synthase. Cancer Res 2000; 60: 2365–2367.[Abstract/Free Full Text]

9. Lai R, Medeiros LJ, Coupland R et al. Immunohistochemical detection of E2F-1 in non-Hodgkin’s lymphomas: a survey of 124 cases. Mod Pathol 1998; 11: 457–463.[ISI][Medline]

10. Salonga D, Danenberg KD, Johnson M et al. Colorectal tumors responding to 5-fluorouracil have low gene expression levels of dihydropyrimidine dehydrogenase, thymidylate synthase, and thymidine phosphorylase. Clin Cancer Res 2000; 6: 1322–1327.[Abstract/Free Full Text]

11. Banerjee D, Schneiders B, Fu JZ et al. Role of E2F-1 in chemosensitivity. Cancer Res 1998; 58: 4292–4296.[Abstract]

12. Wester K, Wahlund E, Sundstrom C et al. Paraffin section storage and immunohistochemistry. Effects of time, temperature, fixation, and retrieval protocol with emphasis on p53 protein and MIB1 antigen. Appl Immunohistochem Mol Morphol 2000; 8: 61–70.[ISI][Medline]

13. Marzio G, Wagener C, Gutierrez MI et al. E2F family members are differentially regulated by reversible acetylation. J Biol Chem 2000; 275: 10887–10892.[Abstract/Free Full Text]

14. Martinez-Balbas MA, Bauer UM, Nielsen SJ et al. Regulation of E2F1 activity by acetylation. EMBO J 2000; 19: 66.





This Article
Abstract
Full Text (PDF)
E-letters: Submit a response
Alert me when this article is cited
Alert me when E-letters are posted
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (1)
Disclaimer
Request Permissions
Google Scholar
Articles by Belvedere, O.
Articles by Sobrero, A.
PubMed
PubMed Citation
Articles by Belvedere, O.
Articles by Sobrero, A.