Abstract

Background: Previously we reported that immunohistochemical examination of p53, bcl-2, glutathione S-transferase-π (GST-π), thymidylate synthase (TS) and vascular endothelial growth factor (VEGF) in biopsy samples was a useful method for predicting clinical outcome of gastric cancer patients treated with 5-fluorouracil and cisplatin. Here, we investigated if these biological markers can predict chemoresponse and survival of unresectable gastric cancer patients treated with irinotecan and cisplatin.

Methods: The subjects were 55 unresectable gastric cancer patients treated with irinotecan (70 mg/m2, Days 1 and 15) and cisplatin (80 mg/m2, Day 1). Expression of p53, bcl-2, VEGF was examined immunohistochemically in biopsy samples.

Results: The overall response rate and the median survival time were 55% (30/55) and 321 days, respectively. Thirty patients with intestinal-type adenocarcinoma survived longer than 25 patients with diffuse-type (median survival time: 446, 259 days, P = 0.013). The favorable phenotypes for chemoresponse were p53-negative, bcl-2-negative and VEGF-positive, which were in accordance with previous findings. The response rate was significantly correlated with the total number of these favorable phenotypes (P = 0.043). The 39 patients having 2 or 3 favorable phenotypes (p53-negative, bcl-2-negative and VEGF-positive) survived longer than the remaining 16 patients (median survival time: 444, 259 days, P = 0.021). In the Cox model, the number of the favorable phenotypes showed a tendency to correlate with survival after adjustment for potentially prognostic factors such as histological type or performance status (P = 0.070).

Conclusions: Immunohistochemical examination of biological markers may be useful in predicting the clinical outcome of unresectable gastric cancer patients treated with irinotecan and cisplatin.

INTRODUCTION

Cisplatin is an active agent against gastric cancer (1), and several chemotherapy regimens including cisplatin have been reported to show high response rates (25). Irinotecan, which inhibits DNA topoisomerase I, is also active against various malignancies including gastric cancer (6). Marked synergism, lack of cross-resistance, different mechanisms of action and relatively different profiles of adverse reactions between irinotecan and cisplatin have encouraged the combination of these agents, and it has shown promising results against gastric and lung cancers. In a phase II study for metastatic gastric cancer, the response rate was 59% and the median survival time was 322 days in 29 patients who had not received previous chemotherapy (7). However, in recent phase III studies, regimens of combined chemotherapy including cisplatin have failed to demonstrate a survival benefit compared with the single agent 5-fluorouracil (5-FU) (810), which had been developed more than 40 years ago (11). The severe toxicity of these cisplatin-containing regimens seems to be one of the reasons for no survival benefit despite the higher response rates as compared with 5-FU alone. It appears that we need to select an optimal regimen for each patient by predicting the chemotherapeutic efficacy. Recently, remarkable advances in the basic research have led to the identification of many biological markers indicative of sensitivity to some antineoplastic agents, some of which have been proved to have clinical impact.

Previously we reported that immunohistochemical examination of biological markers [p53, bcl-2, glutathione S-transferase-π (GST-π), thymidylate synthase (TS) and vascular endothelial growth factor (VEGF)] in biopsy samples was useful method for predicting the effects of chemotherapy. Our report showed that VEGF-positive, TS-negative, p53-negative, GST-π-negative, bcl-2-negative were favorable phenotypes in terms of chemoresponse, and the number of favorable phenotypes was a good indicator of both response and survival in patients with unresectable gastric cancer treated with 5-FU and cisplatin (12). Mutant p53 and bcl-2 proteins protect cancer cells from apoptosis induced by many antineoplastic agents and confer cytotoxic drug resistance (1315). GST-π is an enzyme that plays an important role in cellular detoxification, and increases in this enzyme have been associated with resistance to antineoplastic agents such as CDDP (1618). Because drug delivery is important for the sensitivity of tumors to antineoplastic agents, VEGF may contribute to chemoresponse through the promotion of angiogenesis and/or vascular permeability (19,20). In the present study, we investigated the relationship between immunohistochemical expression of VEGF, p53, bcl-2, GST-π and effects of chemotherapy in patients with unresectable gastric cancer treated with irinotecan and cisplatin. TS was not examined because 5-FU was not included in this regimen.

PATIENTS AND METHODS

Study Population

A total of 55 gastric cancer patients treated with irinotecan and cisplatin were retrospectively included in this exploratory analysis; the study subjects included nine patients entered into a phase II study of combination chemotherapy (7), which is one of experimental arms of ongoing randomized phase III trial in Japan (JCOG 9912), and 46 patients consecutively selected to be suitable for combination chemotherapy using the same regimen in clinical practice at the National Cancer Center Hospital East and its affiliated institutions. All the patients fulfilled the recruitment criteria used in JCOG 9912 except (viii): in brief, (i) histological confirmation of gastric cancer; (ii) Eastern Clinical Oncology Group scale performance status (PS) of 2 or better; (iii) age of 75 years or younger; (iv) no previous chemotherapy; (v) adequate bone marrow, liver and renal function; (vi) no other active malignancy; (vii) no severe medical complication; and (viii) primary tumors from which it was possible to obtain a sufficient amount of cancerous tissue for examining biological markers before chemotherapy.

Treatment Schedule

The treatment schedule of the combination of irinotecan and cisplatin and dose modification were the same as in the phase II study (7), briefly, drip infusion of irinotecan (70 mg/m2, day 1 and 15) and cisplatin (80 mg/m2, day 1) with adequate hydration. This treatment was repeated every 4 weeks until disease progression, patient refusal or unacceptable adverse reactions.

Evaluation of the Effects of Chemotherapy

Responses to chemotherapy in measurable lesions were evaluated by the standard World Health Organization response criteria (21). For primary lesions, responses were evaluated according to the criteria proposed by the Japanese Research Society for Gastric Cancer (22) using either gastroscopy or barium gastrography. Overall response was defined as the sum of the number of complete and partial responses. All patients were followed for at least 1 year after the initiation of chemotherapy, and survival time was defined as the period from the date of initiation of chemotherapy to the date of death due to any cause or the date of last confirmation of survival.

Immunohistochemical Examination

Immunohistochemical staining was performed in the same way as in our previous study (12). All immunohistochemical examinations were performed on tissue sections of formalin-fixed and paraffin-embedded biopsy specimens from primary tumors. Serial 3 μm thick slices were cut, deparaffinized in xylene and dehydrated with a graded series of ethanol solutions, then immersed in methanol containing 0.3% H2O2 for 20 min to inhibit endogenous peroxidase activity. The sections stained for p53 and bcl-2 were heated to 95°C by microwave irradiation for 10 min in phosphate-buffered saline (PBS) and 10 mM citrate buffer, respectively. The sections stained for VEGF were treated with 0.05% pepsin in 0.01 N HCl for 20 min at room temperature. After blocking with 10% normal swine serum in PBS (blocking buffer) for 60 min at room temperature, all sections were incubated overnight at room temperature with the primary antibodies diluted in blocking buffer to the following concentrations: anti-p53 antibody (Nichirei, Tokyo, Japan), 1:20 000; anti-bcl-2 antibody (DAKO, Glostrup, Denmark), 1:40; anti-GST-π antibody (MBL, Nagoya, Japan), 1:24 000; anti-VEGF antibody (Santa Cruz Biochemistry, CA, USA), 1:500. The sections were washed with PBS and then incubated with biotinylated second antibody diluted to 1:200 for 1 h. After washing with PBS, the sections were incubated with ABC reagent (Vector Laboratories, CA, USA), and the color was developed in a reaction mixture containing 2% 3-3′-diaminobenzidine and 0.3% hydrogen peroxide in Tris buffer. The sections were then counterstained with hematoxylin or methyl green. The two investigators, F.N. and N.B., who were blinded to clinical outcome, assessed immunohistochemical staining independently. The intensity of staining of p53 and GST-π was graded as (++) when strong, as (+) when faint and as (−) when no staining was visible. For bcl-2, the intensity of staining was graded as (++) when stronger than that in the case of lymphocytes, as (+) when equal and as (−) when weaker than that in the case of lymphocytes. The staining of VEGF was graded as (+) when the intensity of staining in the case of the cancer cells was stronger than that in the case of stromal cells, as (±) when equal and as (−) when weaker. For all markers, cases were defined as positive when >20% of all cancer cells in each section showed (++) or (+).

Statistical Analysis

Chi-squared test was applied for comparisons between the expression of biological markers and the chemoresponse. Mantel test was applied for comparisons between the chemoresponse and the number of favorable phenotypes. Survival curve was constructed using Kaplan–Meier method and compared using log-rank test. Prognostic importance of the number of favorable phenotypes was analysed using the Cox regression model, which included tumor extension, histological and macroscopic tumor type, performance status, and age as covariates. These covariates were selected because they were recognized as important variables to predict survival in the previous study (12). Statistic analysis was performed by JMP ver. 4.0.5J software (SAS Institute, Inc., Cary, NC, USA).

RESULTS

Patient Backgrounds and Chemotherapeutic Effects

Clinicopathological features are listed in Table 1. The overall response rate was 55% (30/55), and the median survival time (MST) was 321 days. While the response rate did not differ between the patients with intestinal-type and diffuse-type (47%, 56%, P = 0.843), the former survived much longer than the latter (MST: 446, 259 days, P = 0.013).

Table 1.

Clinicopathological features of the subjects

Clinicopathological features
 
No. of patients (%)
 
Sex  
    Male 37 (67) 
    Female 18 (33) 
Performance status  
    0 34 (62) 
    1, 2 21 (38) 
Age (years)  
    >60 27 (49) 
    ≤60 28 (51) 
Macroscopic type  
    Non-scirrhous 20 (36) 
    Scirrhous 35 (64) 
Histological type  
    Intestinal 30 (55) 
    Diffuse 25 (45) 
Degree of tumor extent  
    Locally advanced 20 (36) 
    Metastatic 35 (64) 
Clinicopathological features
 
No. of patients (%)
 
Sex  
    Male 37 (67) 
    Female 18 (33) 
Performance status  
    0 34 (62) 
    1, 2 21 (38) 
Age (years)  
    >60 27 (49) 
    ≤60 28 (51) 
Macroscopic type  
    Non-scirrhous 20 (36) 
    Scirrhous 35 (64) 
Histological type  
    Intestinal 30 (55) 
    Diffuse 25 (45) 
Degree of tumor extent  
    Locally advanced 20 (36) 
    Metastatic 35 (64) 

Biological Marker Expression

Positive staining of p53 was observed in the nuclei of the cancer cells, whereas that of bcl-2 was observed in the cytoplasm. Staining of VEGF was observed in both cancer and stromal cells. The positive rates of p53, bcl-2, VEGF were 44% (24/55), 18% (10/55) and 64% (35/55), respectively. The clinicopathological features did not differ between the expression positive patients and the negative patients for p53, bcl-2 and VEGF, respectively. However, since GST-π-positive patients had a significantly better performance status than GST-π-negative patients, we excluded GST-π from the further investigation.

Biological Marker Expression and Response

Higher response rates were observed in patients with p53 negative, bcl-2 negative and VEGF positive, respectively (Table 2). These relationships were in agreement with the previous findings (12). We, therefore, designated p53-negative, bcl-2-negative and VEGF-positive as favorable phenotypes for chemoresponse.

Table 2.

Expression of biological markers and antitumor response

Marker
 
CR + PR (%)
 
NC + PD (%)
 
Total
 
P-value
 
p53 (−) 20 (65) 11 (35) 31 0.0915 
p53 (+) 10 (42) 14 (58) 24  
bcl-2 (−) 25 (56) 20 (44) 45 0.9999 
bcl-2 (+) 5 (50) 5 (50) 10  
VEGF (+) 22 (63) 13 (37) 35 0.1015 
VEGF (−) 8 (40) 12 (60) 20  
Marker
 
CR + PR (%)
 
NC + PD (%)
 
Total
 
P-value
 
p53 (−) 20 (65) 11 (35) 31 0.0915 
p53 (+) 10 (42) 14 (58) 24  
bcl-2 (−) 25 (56) 20 (44) 45 0.9999 
bcl-2 (+) 5 (50) 5 (50) 10  
VEGF (+) 22 (63) 13 (37) 35 0.1015 
VEGF (−) 8 (40) 12 (60) 20  

CR, complete response; PR, partial response; NC, no change; PD, progressive disease.

Biological Marker Expression and Survival

As a single factor, the patients with favorable phenotypes survived slightly longer than the patients without such favorable phenotypes, but they were not significant (p53, P = 0.504; bcl-2, P = 0.402; VEGF, P = 0.479).

Combination of Biological Markers

The total number of these favorable phenotypes demonstrated a significant association with the response rate (P = 0.043, Table 3). Thirty-nine patients with 2 or 3 favorable phenotypes survived longer than the remaining 16 patients with statistical significance (MST: 444, 259 days, P = 0.021, Fig. 1). Table 4 shows relationships between the number of favorable phenotypes and clinicopathological features which were recognized as important prognostic factors in the previous study. Twenty-five (64%) of the 39 patients with 2 or 3 favorable phenotypes had intestinal-type adenocarcinoma histologically, whereas 5 (31%) of the 16 patients with 1 or 0 had intestinal-type (P = 0.053). There was no difference in other clinicopathological features between the two groups.

Figure 1.

Survival of patients with or without favorable phenotypes. The solid line and dotted line represent patients with or without 2 or 3 favorable phenotypes, respectively (P = 0.021).

Figure 1.

Survival of patients with or without favorable phenotypes. The solid line and dotted line represent patients with or without 2 or 3 favorable phenotypes, respectively (P = 0.021).

Table 3.

Number of favorable phenotypes and antitumor response

No. of favorable phenotypes* Antitumor response
 
  

 
CR + PR (%)
 
NC + PD (%)
 
Total
 
13 (72) 5 (28) 18 
11 (52) 10 (48) 21 
1 or 0 6 (38) 10 (62) 16 
No. of favorable phenotypes* Antitumor response
 
  

 
CR + PR (%)
 
NC + PD (%)
 
Total
 
13 (72) 5 (28) 18 
11 (52) 10 (48) 21 
1 or 0 6 (38) 10 (62) 16 

CR, complete response; PR, partial response; NC, no change; PD, progressive disease.

*Favorable phenotypes include p53-negative, bcl-2-negative and VEGF-positive.

Table 4.

Clinicopathological features and number of favorable phenotypes

Clinicopathological features No. of favorable phenotypes
 
  

 
2 or 3 (%)
 
1 or 0 (%)
 
P-value
 
Age (years); median (range) 60 (26–74) 51 (26–68)  
Sex    
    Male/female 26/13 (67/33) 11/5 (69/31) 0.9999 
Performance status    
    0/1, 2 23/16 (59/41) 11/5 (69/31) 0.7056 
Macroscopic type    
    Non-scirrhous/scirrhous 17/22 (44/56) 3/13 (19/81) 0.1238 
Histological type    
    Intestinal/diffuse 25/14 (64/36) 5/11 (31/69) 0.0537 
Tumor extent    
    Locally advanced/metastatic 13/26 (33/66) 7/9 (44/56) 0.2501 
Clinicopathological features No. of favorable phenotypes
 
  

 
2 or 3 (%)
 
1 or 0 (%)
 
P-value
 
Age (years); median (range) 60 (26–74) 51 (26–68)  
Sex    
    Male/female 26/13 (67/33) 11/5 (69/31) 0.9999 
Performance status    
    0/1, 2 23/16 (59/41) 11/5 (69/31) 0.7056 
Macroscopic type    
    Non-scirrhous/scirrhous 17/22 (44/56) 3/13 (19/81) 0.1238 
Histological type    
    Intestinal/diffuse 25/14 (64/36) 5/11 (31/69) 0.0537 
Tumor extent    
    Locally advanced/metastatic 13/26 (33/66) 7/9 (44/56) 0.2501 

Multivariate Analysis

The covariates in the Cox model were set to the same as those in our previous study to improve the comparability between the previous and present results (Table 5). In the Cox model, the number of the favorable phenotypes showed a tendency to correlate with survival; the prognosis in patients having only <2 favorable phenotypes was poorer compared with that in patients having 2 or 3 favorable phenotypes (hazard ratio: 1.43, P = 0.070). Among the covariates in the model, histological type and performance status were significantly correlated with survival. Tumor extension, macroscopic type and age were not significant in the present study.

Table 5.

Cox proportional regression analysis for survival

Variable
 
Categories
 
P-value
 
Hazard rate ratio (95% CI)
 
No. of favorable phenotypes 2–3 versus 0–1 0.0704 1.433 (0.969–2.104) 
Histological type Intestinal versus diffuse 0.0066 1.695 (1.159–2.487) 
Tumor extension Locally advanced versus metastatic 0.1029 1.346 (0.941–1.942) 
Performance status 0 versus 1 and 2 0.0208 1.514 (1.065–2.173) 
Macroscopic type Non-scirrhous versus scirrhous 0.6078 0.914 (0.651–1.293) 
Age (years) ≤60 versus >60 0.7098 0.942 (0.684–1.287) 
Variable
 
Categories
 
P-value
 
Hazard rate ratio (95% CI)
 
No. of favorable phenotypes 2–3 versus 0–1 0.0704 1.433 (0.969–2.104) 
Histological type Intestinal versus diffuse 0.0066 1.695 (1.159–2.487) 
Tumor extension Locally advanced versus metastatic 0.1029 1.346 (0.941–1.942) 
Performance status 0 versus 1 and 2 0.0208 1.514 (1.065–2.173) 
Macroscopic type Non-scirrhous versus scirrhous 0.6078 0.914 (0.651–1.293) 
Age (years) ≤60 versus >60 0.7098 0.942 (0.684–1.287) 

CI, confidence interval.

DISCUSSION

Our results support the hypothesis that immunohistochemical examination of biological markers may be useful in predicting the clinical outcome for unresectable gastric cancer patients receiving chemotherapy. The number of favorable phenotypes (p53-negative, bcl-2-negative and VEGF-positive) indicates chemotherapeutic effects. We are aware of no published reports that describe relation between biological markers and therapeutic effects in gastric cancer patients treated with irinotecan and cisplatin. Our results also confirm the results of the previous report that suggested the utility of combination of biological markers (12).

Up to the present, a few biological mechanisms have been implicated in determining the sensitivity to antineoplastic agents. Yeh et al. (23) reported that overexpression of p53 was not associated with resistance of gastric cancer to 5-FU-based systemic chemotherapy, whereas Nakata et al. (24) reported that p53 protein overexpression could serve as a predictor of the response to chemotherapy in gastric cancer. Thus, the correlation between some biological markers and chemoresponse is still controversial in cases of gastric cancer.

In the present study, patients who are either p53-negative, bcl-2-negative or VEGF-positive showed only a slightly higher response rate than the others. A single biological marker seems to have a small impact in predicting chemosensitivity, as shown in our previous study. Nakata et al. (25) investigated the relationship between bcl-2 and bax proteins and effect of chemotherapy in gastric cancer patients, and reported that among the bax-positive cases patients with bcl-2-positive tumors were significantly more resistant to 5-FU and had a worse prognosis than bcl-2-negative cases. Some other reports have also described the utility of combination of a couple of biological markers (26,27).

In the present study, the number of favorable phenotypes showed a clear correlation with response rates. Moreover, patients with 2 or 3 favorable phenotypes survived significantly longer than those with 1 or 0. From these results, the number of favorable phenotypes may be a good predictor of therapeutic effects in gastric cancer patients treated with irinotecan and cisplatin. Patients with intestinal-type adenocarcinoma survived longer than those with the diffuse-type, though the reasons for this difference are not clear. Yonemura et al. (28) reported a close relationship between VEGF-C expression, lymphatic spread and prognosis after surgery in gastric cancer patients. During the last few years, it was revealed that VEGF-A plays a role of prime importance in angiogenesis (29,30). The majority of subtypes of VEGF may explain the difference in chemoresponse.

The median survival time of the phase II study of irinotecan and cisplatin was 322 days, and other phase III studies for patients with metastatic gastric cancer generally showed median survival time of 7–11 months (810,31). And the overall response rate and survival in the present study were very similar to the results of phase II study of irinotecan and cisplatin. From these points, the overall survival time of patients with intestinal-type and 2 or 3 favorable phenotypes was remarkably long (data was not shown). Although survival elongation dose not always need high response rate, we already reported usefulness of biological markers and we have a tendency that the number of favorable phenotypes correlates to survival in the present study.

Investigating biological markers with consecutive patients eligible for enrollment criteria based on JCOG9912, as the present study is retrospective and some selection biases may not be excluded, we cannot confirm a utility of combinations of biological markers. We think a utility of biological markers should be confirmed in a large-scale study prospectively. The Japan Clinical Oncology Group had already initiated a three-arm randomized trial comparing 5-FU alone with S-1 alone and with irinotecan and cisplatin (JCOG9912). We are planning to investigate biological markers in this phase III trial.

We are very grateful to Mari Nakane for performing the immunohistochemical staining. This work was supported in part by Grants-in-Aid for Cancer Research (9-3, 11S-3) from the Ministry of Health, Labor and Welfare of Japan.

References

1.
Lacave AJ, Izarzugaza I, Anton Aparicio LM, Valle Pereda M, Gracia Marco JM, Buesa JM. Phase II clinical trial of cis-dichlorodiammineplatinum in gastric cancer.
Am J Clin Oncol
 
1983
;
6
:
35
–8.
2.
Moertel CG, Rubin J, O'Connell MJ, Schutt AJ, Wieand HS. A phase II study of combined 5-fluorouracil, doxorubicin, and cisplatin in the treatment of advanced upper gastrointestinal adenocarcinomas.
J Clin Oncol
 
1986
;
4
:
1053
–7.
3.
Preusser P, Wilke H, Achterrath W, Fink U, Lenaz L, Heinicke A, et al. Phase II study with the combination etoposide, doxorubicin, and cisplatin in advanced measurable gastric cancer.
J Clin Oncol
 
1989
;
7
:
1310
–7.
4.
Findlay M, Cunningham D, Norman A, Mansi J, Nicolson M, Hickish T, et al. A phase II study in advanced gastro-esophageal cancer using epirubicin and cisplatin in combination with continuous infusion 5-fluorouracil (ECF).
Ann Oncol
 
1994
;
5
:
609
–16.
5.
Ohtsu A, Shimada Y, Yoshida S, Saito H, Seki S, Morise K, et al. Phase II study of protracted infusional 5-fluorouracil combined with cisplatinum for advanced gastric cancer: report from the Japan Clinical Oncology Group (JCOG).
Eur J Cancer
 
1994
;
30
:
2091
–3.
6.
Matsuzaki T, Yokokura T, Mutai M, Tsuruo T. Inhibition of spontaneous and experimental metastasis by a new derivative of camptothecin, CPT-11, in mice.
Cancer Chemother Pharmacol
 
1988
;
21
:
308
–12.
7.
Boku N, Ohtsu A, Shimada Y, Shirao K, Seki S, Saito H, et al. Phase II study of a combination of irinotecan and cisplatin against metastatic gastric cancer.
J Clin Oncol
 
1999
;
17
:
319
–23.
8.
Ohtsu A, Shimada Y, Shirao K, Boku N, Hyodo I, Saito H, et al. Randomized phase III trial of fluorouracil alone versus fluorouracil plus cisplatin versus uracil and tegafur plus mitomycin in patients with unresectable, advanced gastric cancer: The Japan Clinical Oncology Group Study (JCOG9205).
J Clin Oncol
 
2003
;
21
:
54
–9.
9.
Cullinan SA, Moertel CG, Wieand HS, O'Connell MJ, Poon MA, Krook JE, et al. Controlled evaluation of three drug combination regimens versus fluorouracil alone for the therapy of advanced gastric cancer.
J Clin Oncol
 
1994
;
12
:
412
–6.
10.
Kim NK, Park YS, Heo DS, Suh C, Kim SY, Park KC, et al. A phase III randomized study of 5-fluorouracil and cisplatin versus 5-fluorouracil, doxorubicin, and mitomycin C versus 5-fluorouracil alone in the treatment of advanced gastric cancer.
Cancer
 
1993
;
71
:
3813
–8.
11.
Heidelberger C, Chandhari NK, Dannenberg P. Fluorinated pyrimidines: a new class of tumor inhibitory compounds.
Nature
 
1957
;
179
:
663
–7.
12.
Boku N, Chin K, Hosokawa K, Ohtsu A, Tajiri H, Yoshida S, et al. Biological markers as a predictor for response and prognosis of unresectable gastric cancer patients treated with 5-fluorouracil and cis-platinum.
Clin Cancer Res
 
1998
;
4
:
1469
–74.
13.
Lowe SW, Ruley HE, Jacks T, Housman DE. p53-dependent apoptosis modulates the cytotoxicity of anticancer agents.
Cell
 
1993
;
74
:
957
–67.
14.
Reed JC, Kitada S, Takayama S, Miyashita T. Regulation of chemoresistance by the bcl-2 oncoprotein in non-Hodgkin's lymphoma and lymphocytic leukemia cell lines.
Ann Oncol
 
1994
;
5
Suppl 1:
61
–5.
15.
Lotem J, Sachs L. Regulation by bcl-2, c-myc, and p53 of susceptibility to induction of apoptosis by heat shock and cancer chemotherapy compounds in differentiation-competent and -defective myeloid leukemic cells.
Cell Growth Differ
 
1993
;
4
:
41
–7.
16.
Ozols RF, O'Dwyer PJ, Hamilton TC, Young RC. The role of glutathione in drug resistance.
Cancer Treat Rev
 
1990
;
17
:
45
–50.
17.
Teicher BA, Holden SA, Kelley MJ, Shea TC, Cucchi CA, Rosowsky A, et al. Characterization of a human squamous carcinoma cell line resistant to cis-diamminedichloroplatinum(II).
Cancer Res
 
1987
;
47
:
388
–93.
18.
Okuyama T, Maehara Y, Endo K, Baba H, Adachi Y, Kuwano M, et al. Expression of glutathione S-transferase-π and sensitivity of human gastric cancer cells to cisplatin.
Cancer
 
1994
;
74
:
1230
–6.
19.
Connolly DT, Heuvelman DM, Nelson R, Olander JV, Eppley BL, Delfino JJ, et al. Tumor vascular permeability factor stimulates endothelial cell growth and angiogenesis.
J Clin Invest
 
1989
;
84
:
1470
–8.
20.
Dvorak HF, Orenstein NS, Carvalho AC, Churchill WH, Dvorak AM, Galli SJ, et al. Induction of a fibrin-gel investment: an early event in line 10 hepatocarcinoma growth mediated by tumor-secreted products.
J Immunol
 
1979
;
122
:
166
–74.
21.
World Health Organization. WHO Handbook for Reporting Results of Cancer Treatment. WHO Offset Publication No. 48, Geneva, Switzerland: World Health Organization
1979
.
22.
Japanese Research Society for Gastric Cancer. Japanese Classification of Gastric Carcinoma. 1st English edn. Tokyo: Kanehara & Co
1995
.
23.
Yeh KH, Shun CT, Chen CL, Lin JT, Lee WJ, Lee PH, et al. Overexpression of p53 is not associated with drug resistance of gastric cancers to 5-fluorouracil-based systemic chemotherapy.
Hepatogastroenterology
 
1999
;
46
:
610
–5.
24.
Nakata B, Chung KH, Ogawa M, Ogawa Y, Yanagawa K, Muguruma K, et al. p53 protein overexpression as a predictor of the response to chemotherapy in gastric cancer.
Surg Today
 
1998
;
28
:
595
–8.
25.
Nakata B, Muguruma K, Hirasawa K, Chung YS, Yamashita Y, Inoue T, et al. Predictive value of Bcl-2 and Bax protein expression for chemotherapeutic effect in gastric cancer. A pilot study.
Oncology
 
1998
;
55
:
543
–7.
26.
Metzger R, Leichman CG, Danenberg KD, Danenberg PV, Lenz HJ, Hayashi K, et al. ERCC1 mRNA levels complement thymidylate synthase mRNA levels in predicting response and survival for gastric cancer patients receiving combination cisplatin and fluorouracil chemotherapy.
J Clin Oncol
 
1998
;
16
:
309
–16.
27.
Tahara M, Ochiai A, Fujimoto J, Boku N, Yasui W, Ohtsu A, et al. Expression of thymidylate synthase, thymidine phosphorylase, dihydropyrimidine dehydrogenase, E2F-1, Bak, Bcl-X, and Bcl-2, and clinical outcomes for gastric cancer patients treated with bolus 5-fluorouracil.
Oncol Rep
 
2004
;
11
:
9
–15.
28.
Yonemura Y, Endo Y, Fujita H, Fushida S, Ninomiya I, Bandou E, et al. Role of vascular endothelial growth factor C expression in the development of lymph node metastasis in gastric cancer.
Clin Cancer Res
 
1999
;
5
:
1823
–9.
29.
Ferrara N, Davis-Smyth T. The biology of vascular endothelial growth factor.
Endocr Rev
 
1997
;
18
:
4
–25.
30.
Ferrara N. Role of vascular endothelial growth factor in the regulation of angiogenesis.
Kidney Int
 
1999
;
56
:
794
–814.
31.
Wils JA, Klein HO, Wagener DJ, Bleiberg H, Reis H, Korsten F, et al. Sequential high-dose methotrexate and fluorouracil combined with doxorubicin—a step ahead in the treatment of advanced gastric cancer: a trial of the European Organization for Research and Treatment of Cancer Gastrointestinal Tract Cooperative Group.
J Clin Oncol
 
1991
;
9
:
827
–31.

Author notes

1Division of Digestive Endoscopy and Gastrointestinal Oncology, National Cancer Center Hospital East, 2Pathology Division, National Cancer Center Research Institute East, Kashiwa, Chiba, 3Department of Internal Medicine, Misawa City Hospital, Aomori, 4Department of Medicine, Yamagata Prefectural Central Hospital, Yamagata, 5Department of Medicine, Saku General Hospital, Nagano, 6Department of Internal Medicine, National Shikoku Cancer Center, Matsuyama and 7Health Service, Kyoto University, Kyoto, Japan