- Email: [email protected]
Is thrombomodulin useful as a tumor marker of a lung cancer?
ELSEVIER
Lung Cancer 15 (1996) 189-195
Is thrombomodulin useful as a tumor marker of a lung cancer? Atsuhisa Tamuraa,*, Hikotaro Komatsua, Akira Hebisawab, Atsuyuki Kurashimaa, Masashi Mori”, Toru Katayamaa “Department “Department
of Respiratory of Pathology,
Diseases, Tokyo
Tokyo National Chest Hospital, 3-I-1, Takeoka, Tokyo 204. Japan National Chest Hospital, 3-I-1, Takeoka, Kiyose, Tokyo
Kiyose, 204. Japan
Received 27 February 1996; revised 8 May 1996; accepted 10 May 1996
Abstract To evaluate whether thrombomodulin (TM) can be considered a tumor marker of lung cancer, we examined serum TM levels in 179 lung cancer patients by an enzyme immunoassay. Further, in 91 resected cases, we also investigated the TM presence in lung cancer tissue by an immunohistochemical method. Our findings revealed that 58 (32%) of 179 lung cancer patients had higher serum TM levels than our cutoff levels, No difference was seen in the serum TM-positive ratio between histologic types, but patients with Stage IV disseminated cancer showed a higher serum TM-positive ratio than those with Stage I-IIIB localized cancer, especially cases of adenocarcinoma. Immunostaining results showed a TM presence in the cancer cells of 32 (35%) of 91 lung cancer cases. The TM presence in cancer cells was higher in cases with squamous cell carcinoma than with adenocarcinoma; no differences were seen among the disease stages. No correlation was found between the serum TM and the presence of TM in cancer cells. Further, serum TM status appears to have little clinical significance as a method to detect early lung cancer or for predicting the recurrence after the surgical excision of a lung cancer. Thus, we have concluded that, although serum TM levels are elevated in some lung cancer patients, it is doubtful that the TM status is useful as a tumor marker of lung cancer.
* Corresponding
author. Tel: + 81 424 912111; fax: + 81 424 942168.
0169-5002/96/$15.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved PIf SO169-5002(95)00582-X
190 Keywords:
A. Tamura
et al. 1 Lung
Lung cancer; Thrombomodulin;
Cancer
15 (1996)
189-195
Serum level; Immunohistochemistry
1. Introduction
Thrombomodulin (TM) is an endothelial glycoprotein that acts as an important modulator of intravascular coagulation [3]. TM has also been found to be identical to fetomodulin, a cell surface protein that modulates embryogenesis and is found on lung bud epithelium [5]. We previously found that functionally active TM is present in some lung cancer cells, which led us to conclude that TM belongs among the oncodevelopmental antigens [12]. TM is also known to be present in soluble form in circulating blood and urine [6]. Further, it has been reported that plasma or serum TM levels increase in patients with generalized endothelial cell injury, such as diabetes mellitus (DM), systemic lupus erythematosus, and/or disseminated intravascular coagulation (DIC); plasma or serum TM levels also increase in patients who manifest renal or hepatic dysfunction due to the reduced catabolism or excretion of TM [2,8,10,11]. The aim of the study is to evaluate the presence of TM in serum and tumor tissue from patients having lung cancer of adenocarcinoma or squamous cell carcinoma and the possible relation of TM to relapse. We have thus investigated whether TM may be useful as a tumor marker of lung cancers, like other oncodevelopmental antigens such as carcinoembryonic antigen.
2. Materials
and methods
Serum samples from 179 lung cancer patients (139 men and 40 women, mean age 61 years) treated between 1987 and 1992 at Tokyo National Chest Hospital were retrospectively studied. Excluded from this study were lung cancer patients with DM, collagen vascular diseases, and DIC (systemic diseases manifesting a generalized endothelial injury). Patients with liver dysfunction and renal dysfunction as evidenced by elevated serum AST or ALT levels and elevated serum creatinine levels, respectively, were also excluded. Also excluded were patients who had received prior therapy and patients in their terminal state. Cases were classified by histologic types and further divided by the stage of the disease [l]. Stages I-IIIA were assigned to resected cases according to pathologic classification of the TNM proven by lobectomy or pneumonectomy, while Stage IIIB and IV were assigned to inoperable cases according to clinical classification of the TNM. Serum samples were collected from each patient before initial treatment, between 08:OO and 10:00 h and stored at - 70°C until analysis. Serum TM immunoreactivity (serum TM level) was measured by an enzyme immunoassay (Mitsubishi Gas Chemical Company Inc., Tokyo, Japan) which was developed by Ishii et al. [7]. Details of the procedure were described elsewhere [ll]. Based on the serum TM
A. Tamura
et al. /Lung
Cancer
15 (1996)
189-195
191
levels (mean + SD.: 16.88 &- 3.26 rig/ml) from 41 healthy controls, our cutoff level to reach 95% specificity was determined to be 23.4 rig/ml. There were no significant inter-assay and intra-assay variations in the immunoassay. In resected cases (Stages I-IIIA), we examined the lung cancer tissue for the presence of TM. Immunohistochemical staining employed the labeled streptavidinbiotin (LSAB) method [4], using an anti-TM antibody (DAKO, Copenhagen, Denmark) and LSAB kit (DAKO, Copenhagen, Denmark); 5-pm-thick, paraffinembedded sections fixed by 10% formalin were used. The sections were pretreated with 0.1% trypsin in 0.05 M Tris-HCl (pH 7.6) containing 5 mM CaCl, at 25°C for 60 min for unmasking of the antigenic sites. The antibody was diluted to 1:50, and the incubation was performed at 4°C overnight. Peroxidase activity was visualized by 3-,3’-diaminobenzidine hydrochloride (20 mg/dl) with 0.05% hydrogen peroxide and by counterstaining the sections with hematoxylin. The specificity of the immunostaining was confirmed by a negative control using nonimmune mouse IgG in place of the primary antibody. The mean values of serum TM obtained in this study were analyzed by Student’s t-test and when the Student’s t-test seemed inadequate, we used Cochran and Cox test. The prevalence of serum and/or tissue TM was evaluated by Chi-square test. P-values of 0.05 or less was considered to be significant.
3. Results
Details of the serum TM findings are shown in Table 1. Fifty-eight (32%) of the 179 patients had serum TM levels that were higher than our cutoff serum TM levels and thus were considered to be serum TM-positive. No difference was noted in the serum TM-positive ratio between the adenocarcinoma and the squamous cell carcinoma patients. It was also noted that adenocarcinoma patients with disseminated disease (Stage IV) had significantly higher serum TM-positive ratios than patients with localized disease (Stages I-IIIB) (P < 0.001, by Chi-square test), whereas in the squamous cell carcinoma patients, no such difference was noted. In 32 (35%) of 91 resected lung cancer cases, TM was present in the lung cancer cells (Fig. 1). TM was mainly present on the cell surface membrane both in adenocarcinoma and squamous cell carcinoma. We considered these cases to be cell TM-positive. Details of cell TM-positive ratios are shown in Table 2. It was also found that the cell TM-positive ratio was higher in the squamous cell carcinoma cases than in the adenocarcinoma cases (P < 0.001, by Chi-square test). Further, while no differences in the cell TM-positive ratio were noted between the stages of the disease, TM-positive cells in the adenocarcinoma cases were present only in Stage I. The relationship between the serum TM-positive cases and cell TM-positive cases was also evaluated, but no correlation was found (Table 3). Finally, no correlation was found between recurrence (mainly distant metastasis) within 2 years after surgery and serum or cell TM-positivity.
A. Tamura
192
et al. 1 Lung
Cancer
IS (1996)
189- 195
4. Discussion
Lindahl et al. [9] have reported that the plasma TM level of colorectal cancer patients increases only in their terminal state. However, our study has found that some lung cancer patients have high serum TM levels not only in disseminated disease but in localized disease. There are two major mechanisms responsible for this serum TM elevation: cancer cells that produce TM and/or an endothelial injury due to a cancer invasion. Thus, it can be speculated that the serum TM-positivity in Stage I lung cancers may be caused by TM-producing cancer cells, and that the higher serum TM-positive ratio seen in Stage IV lung cancers may be largely due to a generalized endothelial injury brought about by metastases, with or without the participation of the TM-producing cancer cells. However, our analysis of these resected cases has not supported these speculations because there was no correlation between serum TM and cancer cell TM, or between serum TM and postsurgical recurrence of a lung cancer. From the clinical viewpoint, the serum TM status appears to have little practical value as a method for detecting a lung cancer, because of the relatively low ratio of serum TM-positive patients with a localized lung cancer. Similarly, the serum TM status cannot be used to predict the recurrence of lung cancer, since no correlation was seen between serum TM-positive cases and recurrence after surgery. Further, although we did not obtain follow-up data of the serum TM status after surgery in Table Serum
1 thrombomodulin
(TM)
findings
Number cases Adenocarcinoma Localized disease Stage I Stage II Stage IIIA Stage IIIB Disseminated disease Stage IV
103 13 29 3 20 21
Squamous cell carcinoma Localized disease Stage I Stage II Stage IIIA Stage IIIB Disseminated disease Stage IV
76 59 21 4 14 20 17
30
in lung cancer
of
.
“Cutoff level, 23.4 rig/ml. *P < 0.001 compared by Cochran and Cox **P < 0.001 compared by Chi-square test.
patients
Serum TM levels + S.D., rig/ml)
Number of TM-positive” cases (%)
21.90 19.76 20.17 20.67 18.19 20.55
+ + k t k i-
27.13
k
10.19*
18 (60)**
22.85 22.00 20.58 25.23 19.85 24.35
k * k + 5 k
9.07 7.71 4.49 8.60 4.71 10.94
26 18 6 1 3 8
25.78
k
12.59
test.
7.55 4.79* 4.96 5.32 3.66 5.40
(mean
32 (31) 14 (19)** 6 (21) 1 (33) 2 (10) 5 (24)
(34) (31) (29) (25) (21) (40)
8 (47)
A. Tamura
et al. /Lung
Cancer
15 (1996)
189-195
193
Fig. 1. Presence of TM in tissue specimens of lung cancers detected by an anti-TM antibody. (A) Positive staining of squamous cell carcinoma cells ( x 270). Staining pattern of TM is punctate along cell membranes. Note positive staining of endothelial cells in the stroma. (B) Negative staining of adenocarcinema cells in contrast to positive staining of endothelial cells ( x 270).
this study, if such data were available, it may show that the TM status can be greatly affected by endothelial cell damage due to the surgery and/or adjuvant radiotherapy or chemotherapy. For these reasons it is felt that the postoperative TM data may be difficult to interpret. Thus, based on these findings and considerations, we have concluded that although serum TM levels are elevated in some lung cancer patients, it appears doubtful that TM status is useful as a tumor marker of lung cancer. Since the progression of cancer is influenced by the coagulation system, TM found in lung cancer cells may modulate the invasiveness and metastatic potential
194
A. Tamura
Table 2 Presence of thrombomodulin
Cancer
15 (1996)
189-195
(TM) in lung cancer cells
Adenocarcinoma Stage I Stage II Stage IIIA Squamous cell carcinoma Stage I Stage II Stage IIIa *P
et al. /Lung
Number of cases
Number of TM-positive
52 29 3 20 39 21 4 14
I 7 0 0 25 14 4 7
cases (%)
(13)* (24) (0) (0) pi)* (67) (100) (50)
K 0.001 compared by Cl&square test.
of these cells by exhibiting anticoagulant activity [12]. In this regard, Tezuka et al. [13] have found that in cases of an esophageal squamous cell carcinoma, TM expression was reduced in the cancer tissue of metastatic lymph nodes in contrast to the primary lesion, which has led them to speculate that this reduction appears to play a role in the metastatic process. As has been noted, TM was frequently found among our patients in the primary lesion of a resected squamous cell carcinoma in all disease stages. In contrast, in resected cases of an adenocarcinoma, although no significant differences were seen between the disease stages, TM was only present in Stage I lung cancers. A possible explanation may be that the loss of TM in the primary lesion of an adenocarcinoma is associated with progression of the disease. Further investigation is required to clearly determine the relationship between the TM presence in cancer cells and the progression of lung cancer.
Table 3 Relationship between serum thrombomodulin Proportion -
(0%)
(TM) and TM stained lung cancer cells of TM stained cancer cells l+ (
2+ (lo-50%)
3t
Adenocarcinoma Serum TM positive” cases Serum TM negative cases
45 I 38
6 2 4
1 0 1
0 0 0
Squamous cell carcinoma Serum TM positive” cases Serum TM negative cases
14 2 12
10 3 I
I 3 4
8 2 6
“Over cutoff serum levels of TM.
(>SO%)
A. Tamura
et al. /Lung
Cancer
15 (1996)
195
189-195
Acknowledgements The enzyme immunoassay system was kindly Chemical Company Inc., Tokyo, Japan.
provided
by Mitsubishi
Gas
References [I] American Joint Committee on Cancer (AJCC). Lung. In: Beahrs OH, Henson DE, Hutter RVP, Kennedy BJ, editors. Manual for staging of cancer, 4th edn. Philadelphia, PA: Lippincott, 1992; 115-122. [2] Boehme MWJ, Nawroth PP, Kling E et al. Serum thrombomodulin: a novel marker of disease activity in systemic lupus erythematosus. Arthritis Rheum 1994; 37: 572-577. [3] Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation. J Biol Chem 1989; 264: 4743-4746. [4] Guesdon JL, Ternynck T, Avrameas S. The use of avidin-biotin interaction in immunoenzymatic techniques. J Histochem Cytochem 1979; 27: 113 1- 1139. [5] Imada S, Yamaguchi H, Nagumo M, Katayanagi S, Iwasaki H, Imada M. Identification of fetomodulin, a surface marker protein of fetal development, as thrombomodulin by gene cloning and functional assays. Dev Biol 1990; 140: 113- 122. [6] Ishii H, Majerus PW. Thrombomodulin is present in human plasma and urine. J Clin Invest 1985; 76: 2178-2181. [7] Ishii H, Nakano M, Tsubouchi J et al. Establishment of enzyme immunoassay of human thrombomodulin in plasma and urine using monoclonal antibodies. Thromb Haemost 1990; 63: 157-162. [8] Iwashima Y, Sato T, Watanabe K et al. Elevation of plasma thrombomodulin level in diabetic patients with early diabetic nephropathy. Diabetes 1990; 39: 983-988. [9] Lindahl AK, Boffa MC, Abildgaard U. Increased plasma thrombomodulin in cancer patients. Thromb Haemost 1993; 69: 112-114. [lo] Takano S, Kimura S, Ohdama S, Aoki N. Plasma thrombomodulin in health and diseases. Blood 1990; 76: 2024-2029. [I I] Takaya M, Ichikawa Y, Kobayashi N et al. Serum thrombomodulin and anticardiolipin antibodies in patients with systemic lupus erythematosus. Clin Exp Rheumatol 1991; 9: 4955499. [12] Tamura A, Matsubara 0, Hirokawa K, Aoki N. Detection of thrombomodulin in human lung cancer cells. Am J Path01 1993; 142: 79-85. [13] Tezuka Y, Yonezawa S, Maruyama I et al. Expression of thrombomodulin in esophageal squamous cell carcinoma and its relationship to lymph node metastasis. Cancer Res 1995; 55: 4196-4200.
Lung Cancer 15 (1996) 189-195
Is thrombomodulin useful as a tumor marker of a lung cancer? Atsuhisa Tamuraa,*, Hikotaro Komatsua, Akira Hebisawab, Atsuyuki Kurashimaa, Masashi Mori”, Toru Katayamaa “Department “Department
of Respiratory of Pathology,
Diseases, Tokyo
Tokyo National Chest Hospital, 3-I-1, Takeoka, Tokyo 204. Japan National Chest Hospital, 3-I-1, Takeoka, Kiyose, Tokyo
Kiyose, 204. Japan
Received 27 February 1996; revised 8 May 1996; accepted 10 May 1996
Abstract To evaluate whether thrombomodulin (TM) can be considered a tumor marker of lung cancer, we examined serum TM levels in 179 lung cancer patients by an enzyme immunoassay. Further, in 91 resected cases, we also investigated the TM presence in lung cancer tissue by an immunohistochemical method. Our findings revealed that 58 (32%) of 179 lung cancer patients had higher serum TM levels than our cutoff levels, No difference was seen in the serum TM-positive ratio between histologic types, but patients with Stage IV disseminated cancer showed a higher serum TM-positive ratio than those with Stage I-IIIB localized cancer, especially cases of adenocarcinoma. Immunostaining results showed a TM presence in the cancer cells of 32 (35%) of 91 lung cancer cases. The TM presence in cancer cells was higher in cases with squamous cell carcinoma than with adenocarcinoma; no differences were seen among the disease stages. No correlation was found between the serum TM and the presence of TM in cancer cells. Further, serum TM status appears to have little clinical significance as a method to detect early lung cancer or for predicting the recurrence after the surgical excision of a lung cancer. Thus, we have concluded that, although serum TM levels are elevated in some lung cancer patients, it is doubtful that the TM status is useful as a tumor marker of lung cancer.
* Corresponding
author. Tel: + 81 424 912111; fax: + 81 424 942168.
0169-5002/96/$15.00 0 1996 Elsevier Science Ireland Ltd. All rights reserved PIf SO169-5002(95)00582-X
190 Keywords:
A. Tamura
et al. 1 Lung
Lung cancer; Thrombomodulin;
Cancer
15 (1996)
189-195
Serum level; Immunohistochemistry
1. Introduction
Thrombomodulin (TM) is an endothelial glycoprotein that acts as an important modulator of intravascular coagulation [3]. TM has also been found to be identical to fetomodulin, a cell surface protein that modulates embryogenesis and is found on lung bud epithelium [5]. We previously found that functionally active TM is present in some lung cancer cells, which led us to conclude that TM belongs among the oncodevelopmental antigens [12]. TM is also known to be present in soluble form in circulating blood and urine [6]. Further, it has been reported that plasma or serum TM levels increase in patients with generalized endothelial cell injury, such as diabetes mellitus (DM), systemic lupus erythematosus, and/or disseminated intravascular coagulation (DIC); plasma or serum TM levels also increase in patients who manifest renal or hepatic dysfunction due to the reduced catabolism or excretion of TM [2,8,10,11]. The aim of the study is to evaluate the presence of TM in serum and tumor tissue from patients having lung cancer of adenocarcinoma or squamous cell carcinoma and the possible relation of TM to relapse. We have thus investigated whether TM may be useful as a tumor marker of lung cancers, like other oncodevelopmental antigens such as carcinoembryonic antigen.
2. Materials
and methods
Serum samples from 179 lung cancer patients (139 men and 40 women, mean age 61 years) treated between 1987 and 1992 at Tokyo National Chest Hospital were retrospectively studied. Excluded from this study were lung cancer patients with DM, collagen vascular diseases, and DIC (systemic diseases manifesting a generalized endothelial injury). Patients with liver dysfunction and renal dysfunction as evidenced by elevated serum AST or ALT levels and elevated serum creatinine levels, respectively, were also excluded. Also excluded were patients who had received prior therapy and patients in their terminal state. Cases were classified by histologic types and further divided by the stage of the disease [l]. Stages I-IIIA were assigned to resected cases according to pathologic classification of the TNM proven by lobectomy or pneumonectomy, while Stage IIIB and IV were assigned to inoperable cases according to clinical classification of the TNM. Serum samples were collected from each patient before initial treatment, between 08:OO and 10:00 h and stored at - 70°C until analysis. Serum TM immunoreactivity (serum TM level) was measured by an enzyme immunoassay (Mitsubishi Gas Chemical Company Inc., Tokyo, Japan) which was developed by Ishii et al. [7]. Details of the procedure were described elsewhere [ll]. Based on the serum TM
A. Tamura
et al. /Lung
Cancer
15 (1996)
189-195
191
levels (mean + SD.: 16.88 &- 3.26 rig/ml) from 41 healthy controls, our cutoff level to reach 95% specificity was determined to be 23.4 rig/ml. There were no significant inter-assay and intra-assay variations in the immunoassay. In resected cases (Stages I-IIIA), we examined the lung cancer tissue for the presence of TM. Immunohistochemical staining employed the labeled streptavidinbiotin (LSAB) method [4], using an anti-TM antibody (DAKO, Copenhagen, Denmark) and LSAB kit (DAKO, Copenhagen, Denmark); 5-pm-thick, paraffinembedded sections fixed by 10% formalin were used. The sections were pretreated with 0.1% trypsin in 0.05 M Tris-HCl (pH 7.6) containing 5 mM CaCl, at 25°C for 60 min for unmasking of the antigenic sites. The antibody was diluted to 1:50, and the incubation was performed at 4°C overnight. Peroxidase activity was visualized by 3-,3’-diaminobenzidine hydrochloride (20 mg/dl) with 0.05% hydrogen peroxide and by counterstaining the sections with hematoxylin. The specificity of the immunostaining was confirmed by a negative control using nonimmune mouse IgG in place of the primary antibody. The mean values of serum TM obtained in this study were analyzed by Student’s t-test and when the Student’s t-test seemed inadequate, we used Cochran and Cox test. The prevalence of serum and/or tissue TM was evaluated by Chi-square test. P-values of 0.05 or less was considered to be significant.
3. Results
Details of the serum TM findings are shown in Table 1. Fifty-eight (32%) of the 179 patients had serum TM levels that were higher than our cutoff serum TM levels and thus were considered to be serum TM-positive. No difference was noted in the serum TM-positive ratio between the adenocarcinoma and the squamous cell carcinoma patients. It was also noted that adenocarcinoma patients with disseminated disease (Stage IV) had significantly higher serum TM-positive ratios than patients with localized disease (Stages I-IIIB) (P < 0.001, by Chi-square test), whereas in the squamous cell carcinoma patients, no such difference was noted. In 32 (35%) of 91 resected lung cancer cases, TM was present in the lung cancer cells (Fig. 1). TM was mainly present on the cell surface membrane both in adenocarcinoma and squamous cell carcinoma. We considered these cases to be cell TM-positive. Details of cell TM-positive ratios are shown in Table 2. It was also found that the cell TM-positive ratio was higher in the squamous cell carcinoma cases than in the adenocarcinoma cases (P < 0.001, by Chi-square test). Further, while no differences in the cell TM-positive ratio were noted between the stages of the disease, TM-positive cells in the adenocarcinoma cases were present only in Stage I. The relationship between the serum TM-positive cases and cell TM-positive cases was also evaluated, but no correlation was found (Table 3). Finally, no correlation was found between recurrence (mainly distant metastasis) within 2 years after surgery and serum or cell TM-positivity.
A. Tamura
192
et al. 1 Lung
Cancer
IS (1996)
189- 195
4. Discussion
Lindahl et al. [9] have reported that the plasma TM level of colorectal cancer patients increases only in their terminal state. However, our study has found that some lung cancer patients have high serum TM levels not only in disseminated disease but in localized disease. There are two major mechanisms responsible for this serum TM elevation: cancer cells that produce TM and/or an endothelial injury due to a cancer invasion. Thus, it can be speculated that the serum TM-positivity in Stage I lung cancers may be caused by TM-producing cancer cells, and that the higher serum TM-positive ratio seen in Stage IV lung cancers may be largely due to a generalized endothelial injury brought about by metastases, with or without the participation of the TM-producing cancer cells. However, our analysis of these resected cases has not supported these speculations because there was no correlation between serum TM and cancer cell TM, or between serum TM and postsurgical recurrence of a lung cancer. From the clinical viewpoint, the serum TM status appears to have little practical value as a method for detecting a lung cancer, because of the relatively low ratio of serum TM-positive patients with a localized lung cancer. Similarly, the serum TM status cannot be used to predict the recurrence of lung cancer, since no correlation was seen between serum TM-positive cases and recurrence after surgery. Further, although we did not obtain follow-up data of the serum TM status after surgery in Table Serum
1 thrombomodulin
(TM)
findings
Number cases Adenocarcinoma Localized disease Stage I Stage II Stage IIIA Stage IIIB Disseminated disease Stage IV
103 13 29 3 20 21
Squamous cell carcinoma Localized disease Stage I Stage II Stage IIIA Stage IIIB Disseminated disease Stage IV
76 59 21 4 14 20 17
30
in lung cancer
of
.
“Cutoff level, 23.4 rig/ml. *P < 0.001 compared by Cochran and Cox **P < 0.001 compared by Chi-square test.
patients
Serum TM levels + S.D., rig/ml)
Number of TM-positive” cases (%)
21.90 19.76 20.17 20.67 18.19 20.55
+ + k t k i-
27.13
k
10.19*
18 (60)**
22.85 22.00 20.58 25.23 19.85 24.35
k * k + 5 k
9.07 7.71 4.49 8.60 4.71 10.94
26 18 6 1 3 8
25.78
k
12.59
test.
7.55 4.79* 4.96 5.32 3.66 5.40
(mean
32 (31) 14 (19)** 6 (21) 1 (33) 2 (10) 5 (24)
(34) (31) (29) (25) (21) (40)
8 (47)
A. Tamura
et al. /Lung
Cancer
15 (1996)
189-195
193
Fig. 1. Presence of TM in tissue specimens of lung cancers detected by an anti-TM antibody. (A) Positive staining of squamous cell carcinoma cells ( x 270). Staining pattern of TM is punctate along cell membranes. Note positive staining of endothelial cells in the stroma. (B) Negative staining of adenocarcinema cells in contrast to positive staining of endothelial cells ( x 270).
this study, if such data were available, it may show that the TM status can be greatly affected by endothelial cell damage due to the surgery and/or adjuvant radiotherapy or chemotherapy. For these reasons it is felt that the postoperative TM data may be difficult to interpret. Thus, based on these findings and considerations, we have concluded that although serum TM levels are elevated in some lung cancer patients, it appears doubtful that TM status is useful as a tumor marker of lung cancer. Since the progression of cancer is influenced by the coagulation system, TM found in lung cancer cells may modulate the invasiveness and metastatic potential
194
A. Tamura
Table 2 Presence of thrombomodulin
Cancer
15 (1996)
189-195
(TM) in lung cancer cells
Adenocarcinoma Stage I Stage II Stage IIIA Squamous cell carcinoma Stage I Stage II Stage IIIa *P
et al. /Lung
Number of cases
Number of TM-positive
52 29 3 20 39 21 4 14
I 7 0 0 25 14 4 7
cases (%)
(13)* (24) (0) (0) pi)* (67) (100) (50)
K 0.001 compared by Cl&square test.
of these cells by exhibiting anticoagulant activity [12]. In this regard, Tezuka et al. [13] have found that in cases of an esophageal squamous cell carcinoma, TM expression was reduced in the cancer tissue of metastatic lymph nodes in contrast to the primary lesion, which has led them to speculate that this reduction appears to play a role in the metastatic process. As has been noted, TM was frequently found among our patients in the primary lesion of a resected squamous cell carcinoma in all disease stages. In contrast, in resected cases of an adenocarcinoma, although no significant differences were seen between the disease stages, TM was only present in Stage I lung cancers. A possible explanation may be that the loss of TM in the primary lesion of an adenocarcinoma is associated with progression of the disease. Further investigation is required to clearly determine the relationship between the TM presence in cancer cells and the progression of lung cancer.
Table 3 Relationship between serum thrombomodulin Proportion -
(0%)
(TM) and TM stained lung cancer cells of TM stained cancer cells l+ (
2+ (lo-50%)
3t
Adenocarcinoma Serum TM positive” cases Serum TM negative cases
45 I 38
6 2 4
1 0 1
0 0 0
Squamous cell carcinoma Serum TM positive” cases Serum TM negative cases
14 2 12
10 3 I
I 3 4
8 2 6
“Over cutoff serum levels of TM.
(>SO%)
A. Tamura
et al. /Lung
Cancer
15 (1996)
195
189-195
Acknowledgements The enzyme immunoassay system was kindly Chemical Company Inc., Tokyo, Japan.
provided
by Mitsubishi
Gas
References [I] American Joint Committee on Cancer (AJCC). Lung. In: Beahrs OH, Henson DE, Hutter RVP, Kennedy BJ, editors. Manual for staging of cancer, 4th edn. Philadelphia, PA: Lippincott, 1992; 115-122. [2] Boehme MWJ, Nawroth PP, Kling E et al. Serum thrombomodulin: a novel marker of disease activity in systemic lupus erythematosus. Arthritis Rheum 1994; 37: 572-577. [3] Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation. J Biol Chem 1989; 264: 4743-4746. [4] Guesdon JL, Ternynck T, Avrameas S. The use of avidin-biotin interaction in immunoenzymatic techniques. J Histochem Cytochem 1979; 27: 113 1- 1139. [5] Imada S, Yamaguchi H, Nagumo M, Katayanagi S, Iwasaki H, Imada M. Identification of fetomodulin, a surface marker protein of fetal development, as thrombomodulin by gene cloning and functional assays. Dev Biol 1990; 140: 113- 122. [6] Ishii H, Majerus PW. Thrombomodulin is present in human plasma and urine. J Clin Invest 1985; 76: 2178-2181. [7] Ishii H, Nakano M, Tsubouchi J et al. Establishment of enzyme immunoassay of human thrombomodulin in plasma and urine using monoclonal antibodies. Thromb Haemost 1990; 63: 157-162. [8] Iwashima Y, Sato T, Watanabe K et al. Elevation of plasma thrombomodulin level in diabetic patients with early diabetic nephropathy. Diabetes 1990; 39: 983-988. [9] Lindahl AK, Boffa MC, Abildgaard U. Increased plasma thrombomodulin in cancer patients. Thromb Haemost 1993; 69: 112-114. [lo] Takano S, Kimura S, Ohdama S, Aoki N. Plasma thrombomodulin in health and diseases. Blood 1990; 76: 2024-2029. [I I] Takaya M, Ichikawa Y, Kobayashi N et al. Serum thrombomodulin and anticardiolipin antibodies in patients with systemic lupus erythematosus. Clin Exp Rheumatol 1991; 9: 4955499. [12] Tamura A, Matsubara 0, Hirokawa K, Aoki N. Detection of thrombomodulin in human lung cancer cells. Am J Path01 1993; 142: 79-85. [13] Tezuka Y, Yonezawa S, Maruyama I et al. Expression of thrombomodulin in esophageal squamous cell carcinoma and its relationship to lymph node metastasis. Cancer Res 1995; 55: 4196-4200.