Ann Thorac Surg 2000;70:1634-1640
© 2000 The Society of Thoracic Surgeons
Original articles: general thoracic
Accelerated growth signals and low tumor-infiltrating lymphocyte levels predict poor outcome in T4 esophageal squamous cell carcinoma
Masahiro Yasunaga, MDa,
Yoichi Tabira, MDa,
Kanyu Nakano, MDa,
Shinichi Iida, MDa,
Nobuuki Ichimaru, MDa,
Nobuhide Nagamoto, MDa,
Tomonori Sakaguchi, MDa
a First Department of Surgery, School of Medicine, Kumamoto University, Kumamoto, Japan
Address reprint requests to Dr Yasunaga, First Department of Surgery, School of Medicine, Kumamoto University, Honjo 111, Kumamoto, Japan
 |
Abstract
|
|---|
Background. Little is known about the biological nature of T4 esophageal carcinoma growth signals and host defenses.
Methods. Paraffin-embedded sections from 78 patients with T2 to T4 esophageal squamous cell carcinoma who underwent operation were analyzed with immunohistochemistry.
Results. Positive cyclin A showed a significantly greater increase in T4 tumors than in those of other stages, and negative p27 showed a significantly greater decrease in T4 tumors than in large T3 stage tumors (tumor size 
4.0 cm). Patients with low-grade tumor-infiltrating lymphocyte (TIL) density showed a significantly greater decrease in T4 than in T2. The combination of p27 and cyclin A was a significant independent prognostic factor among T and N factors in multivariate analysis. TIL density was an independent prognostic factor among immunonutritional variables such as serum albumin concentration and the number of total blood lymphocytes.
Conclusions. T4 esophageal squamous cell carcinoma has a poor prognosis, which is associated with increased p27-negative and cyclin A-positive growth signals in the tumor and with low TIL density in the host.
|
Introduction
|
|---|
The survival rate for patients with esophageal carcinoma has improved due to the development of surgical techniques, the introduction of chemoradiotherapy, and early detection [1–6]. However, advanced carcinoma continues to have a poor prognosis [1–5]. The extent of tumor infiltration (T stage) and the presence of metastatic lymph nodes (N stage) are important factors that determine prognosis [3, 7]. The prognosis of T4 carcinomas, most of which are not resectable, is poor with a mean survival time of less than 1 year [3, 7]. Surgical procedures are reported to worsen the outcome in the majority of microscopically or macroscopically irradical resections [3, 7]. T4 carcinoma is believed to involve strong growth activity, high metastatic potential, and increased cell immortality. Moreover, catabolism caused by dysphagia and weight loss weaken host defenses, which permit tumor progression. However, the biological nature of T4 carcinoma and host defenses on the course T4 carcinoma has not been reported extensively. Further study is necessary to elucidate the relationships and to develop new therapies and individual management protocols. In addition to tumor and host factors, clarification is needed as to how surgical procedures for T4 carcinoma affect prognosis.
In this study, we examined p27, proliferating cell nuclear antigen (PCNA), and cyclin A as cell cycle markers, the amounts of which depend on the degree of the growth signals and which are independent of oncogenic overexpression [8, 9]. We also examined the apoptotic index (AI) to evaluate cell immortality and tumor-infiltrating lymphocyte (TIL) density to evaluate host defense activity [10, 11]. These biological markers can predict the course of advanced esophageal carcinoma beyond conventional prognostic indicators and immunonutritional indicators such as T or N stage, serum albumin concentration, and the total blood lymphocyte count.
|
Patients and methods
|
|---|
Patients
Patients with primary squamous cell carcinoma of the esophagus who underwent esophagectomy with thoracotomy for radical lymphadenectomy at the First Department of Surgery, Kumamoto University Hospital, between 1984 and 1995, were classified into four groups according to tumor stage: T2 stage (17 patients), small T3 stage (tumor size < 4 cm; 18 patients), large T3 stage (tumor size 
BORDER="0"> 4 cm; 22 patients), and T4 stage (21 patients). This classification made biological features including heterogeneity and developmental process equal between the large T3 group and the T4 group. Twenty-eight patients received adjuvant chemotherapy including cisplatin and 5-fluorouracil; these patients were not excluded because their background factors including prognosis were statistically equivalent to those for patients who did not receive adjuvant treatment.
Immunohistochemistry
Tumor samples were fixed in 10% formalin and embedded in paraffin. Two or three representative blocks that included the greatest dimension of tumor were selected and sectioned for subsequent immunostaining. Sections were dewaxed, rehydrated, treated with 3% hydrogen peroxide (Wako, Osaka, Japan) in distilled water to inhibit endogenous peroxidase activity, and microwaved in 10 mM citrate buffer (pH 6.0) for antigen retrieval. Sections were washed with phosphate-buffered saline (PBS) (pH 7.4) and treated with bovine serum albumin (Sigma, St. Louis, MO) for 10 minutes. Slides were incubated with either monoclonal anti-PCNA antibody (PC10, DAKO, Glostrup, Denmark) diluted 1:100, monoclonal anti-p27 antibody (F-8, Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:10, or polyclonal anti-cyclin A antibody (H-432, Santa Cruz Biotechnology) diluted 1:20, then washed with PBS. Anti-mouse Ig horseradish peroxidase-linked F(ab')2 fragment (Amersham, Buckinghamshire, England) or anti-rabbit Ig horseradish peroxidase-linked F(ab')2 fragment (Amersham) was used as secondary antibodies. Reaction products of peroxidase were visualized by incubation with 0.05 mol/L Tris-HCl buffer (pH 7.6) containing 20 mg 3,3'-diaminobenzidine (Dojindo, Kumamoto, Japan) and 20 mL H2O2/100 mL of buffer, then lightly counterstained with hematoxylin. Sections from human tonsil served as positive controls for PCNA and cyclin A.
Negative controls were obtained with mouse normal IgG1 and IgG2a, and rabbit normal immunoglobulin fraction (DAKO) at the same dilutions as the primary antibodies.
The percentage of positive cells was calculated by counting more than 10,000 cells in random high-power fields (10 x 40).
In situ detection of apoptosis
With an in situ apoptosis detection kit (Oncor, Gaithersburg, MD), paraffin-embedded sections were deparaffinized and then incubated with proteinase K (20 g/mL) in PBS for 15 minutes at room temperature, washed in PBS, and then incubated in PBS containing 3% hydrogen peroxide.
Apoptotic index (percentage of apoptotic cells) was determined by counting cells in tumor areas in several randomly chosen sections.
TIL score
Tumor-infiltrating lymphocyte counts were expressed as a percentage of small mononuclear cells seen among cells in hematoxylin and eosin-stained sections. Tumor-infiltrating lymphocyte density was graded as dense, moderate, or weak according to previous reports [12, 13].
Statistical analyses
The 
2 test and analysis of variance (ANOVA) test were used to compare differences in selected features between the four groups. Survival was estimated by the Kaplan-Meier method, and statistical differences were analyzed with the log-rank test. The Cox proportional hazards model was used for multivariate survival analysis. A p value less than 0.05 was considered statistically significant.
|
Results
|
|---|
Demographics
In the current study no significant differences were observed between the four groups with respect to age, sex, lymph node metastasis, histologic grade, lymphatic invasion, vascular invasion, or total lymphocyte count.
Mean tumor size of the T4 group was significantly larger than that of the small T3 group and the T2 group but smaller than that of the large T3 group. Serum total protein concentration of the T4 group was significantly lower than that of the small T3 group. Serum albumin concentration of the T4 group was significantly lower than that of the small T3 group and the large T3 group (Table 1).
The cumulative 5-year survival curves shown in Figure 1 indicate that the T4 group had a significantly poorer prognosis than the other groups according to log-rank test results.
View larger version (17K):
[in this window]
[in a new window]
|
Fig 1. Kaplan-Meier plot for cumulative survival in 78 patients with esophageal cancer. Comparison between patients with T2 stage (n = 17), small T3 stage (n = 18), large T3 stage (n = 22), and T4 stage (n = 21).
|
|
Surgical features
The proportion of noncomplete resections was significantly higher in the T4 group than in other groups because of the positive surgical margin of the invaded organ. Therefore, local recurrence was increased in the T4 group and differed significantly between the T2 group and the T4 group. No significant differences were observed between the four groups with respect to complication morbidity and mortality (Table 2).
Biological features
Mean p27 expression was lower in the T4 group than in the T2 (p = 0.06), small T3 (p = 0.1), and large T3 (p < 0.05) groups. Mean PCNA was not significantly different between the groups, although the T4 group tended to have a higher PCNA than that of the T2 and small T3 groups and a lower value than that of the large T3 group. Mean cyclin A expression was significantly higher in the T4 group than in the other groups. Mean tumor AI was lower in the T4 group than in the T2 group (p = 0.09). AI was also lower in the T4 group than in the small and large T3 groups (p = 0.2 and p = 0.3, respectively). The distribution of TIL density grades was significantly different between the T2 and small T3 groups and between the T2 and T4 groups (Table 3).
Relation between prognosis and combination of p27 and cyclin A
Among the tumor biological markers, the correlation of survival and combination of p27 and cyclin A, which was significantly elevated in the T4 group, was examined. Patients showing negative p27 (< 10%) and positive cyclin A ( 10%) (n = 18) had a 0% 5-year survival rate. Those showing negative p27 and negative cyclin A (< 10%) (n = 15) had a 20% 5-year survival rate. Those showing positive p27 ( 10%) and positive cyclin A (n = 32) had a 26.1% 5-year survival rate. Those showing positive p27 and negative cyclin A (n = 13) had a 64.1% 5-year survival rate. Negative p27 and positive cyclin A was negatively correlated with a significantly poorer prognosis (p < 0.01) (Fig 2).
View larger version (22K):
[in this window]
[in a new window]
|
Fig 2. Kaplan-Meier plot for cumulative survival in 78 patients with esophageal cancer. Comparison between patients with combined positive p27 and negative cyclin A (n = 13), those with positive p27 and positive cyclin A (n = 32), those with negative p27 and negative cyclin A (n = 15), and those with negative p27 and positive cyclin A (n = 18).
|
|
Relation between prognosis and TIL density
The correlation between survival and TIL density, which is related to host defenses, was investigated. Patients showing low-grade TIL density (n = 45) had a 19.5% 5-year survival rate. Those showing high- and moderate-grade TIL density (n = 33) had a 36.1% survival rate. Those showing low TIL density tended to have a worse prognosis (p = 0.7) (Fig 3).
View larger version (16K):
[in this window]
[in a new window]
|
Fig 3. Kaplan-Meier plot for cumulative survival in 78 patients with esophageal cancer. Comparison between patients with high and moderate TIL density (n = 33) and those with low TIL density (n = 45).
|
|
Multivariate analysis of prognostic factors
The effect of the combination of p27 and cyclin A was examined by multivariate analysis together with pathologic T classification and lymph node metastasis as tumor variables. The combination of p27 and cyclin A was shown to be a significant independent factor. Similarly, the effect of TIL density on prognosis was examined by multivariate analysis together with other classic immunonutritional features (serum total protein, serum albumin, and total blood lymphocyte count) as host variables. TIL density was shown to be a significant independent factor (Table 4).
|
Comment
|
|---|
T4 esophageal carcinoma has a poor prognosis [1–6]. The extent of tumor infiltration (T stage) and the presence of metastatic lymph nodes (N stage) are important features that determine prognosis [3, 7]. However, these conventional factors are insufficient for understanding the biological nature of advanced carcinoma and do not include host defense. It is necessary to find biological markers of both tumor and host defenses that can predict the course of T4 carcinoma. Moreover, there is a possibility that surgical procedures for T4 carcinoma worsen outcome given the high operative morbidity and mortality. Analysis of biological markers associated with tumor and host defenses, in conjunction with the influence of surgical procedures in this type of carcinoma, need to be clarified. In this study we examined features of T4 carcinoma in relation to conventional clinicopathologic and immunonutritional findings, surgical features, and biological features, the understanding of which yields useful guidelines for cure-oriented treatment of esophageal carcinoma.
In relation to surgical features, the rate of complete resection of T4 carcinoma was poor in our study, which leads to a higher risk for local recurrence. Additionally, complete resection in this study did not improve the survival rate of those with T4 carcinoma. Although complete resection suppressed local recurrence in our series, lymphatic and hematogenous recurrences occurred. Rice and coworkers [7] reported that T stage was an important predictor of N stage and that T3 and T4 carcinomas frequently spread to regional lymph nodes, which worsened outcome. Although T4 carcinoma in our study was not associated with a high occurrence of positive nodal metastases, the presence of regional lymph node metastases (data not shown), including distant nodal metastases that were not dissected, should show a positive correlation between T4 carcinoma and nodal metastases. T4 carcinoma is a systemic disease and cannot be treated by operation alone [1–4]. Although operative morbidity and mortality in T4 carcinoma occurred at essentially an equivalent incident rate as that for the other groups in our study, indication of operation for T4 carcinoma should be decided carefully.
In conventional pathologic findings such as histologic grade, lymphatic or venous invasion did not sufficiently explain the nature of T4 carcinoma. Serum total protein concentration and albumin concentration in T4 carcinoma were lower than those in the other stage groups, which was explained by an increased catabolism in T4 carcinoma caused by dysphagia and weight loss. However, total blood lymphocyte count was not decreased in T4 compared with the other stage groups. Lymphopoiesis was not disturbed. Another immunologic factor must explain the features of T4 carcinoma.
In relation to biological features, several studies have reported that cell cycle markers are related to prognosis in patients with esophageal carcinoma [14–16]. Increased PCNA and positive cyclin A was reported to be associated with poor prognosis and was observed in the highly advanced stage of esophageal carcinoma [15, 16]. Negative p27 was also reported to be associated with poor prognosis in the early stage of esophageal carcinoma, but other upper digestive carcinomas were reported to be associated with both poor prognosis and advanced staging [14, 17, 18]. In these previous reports, these biological markers were not examined in relation to their specific association with T4 carcinoma [14–18]. In our study, T4 carcinoma was associated with low p27 expression and high cyclin A expression but not with the degree of PCNA expression, which indicates that accelerated growth signals occur mainly during the G1 and G2/M phases. In the natural world, various types of living cells, including bacteria, grow in a cell cycle-dependent manner, and growth signals of G1 and G2 increase more in a comfortable environment than in an uncomfortable one, which causes high-speed turnover of cell cycles [19]. These results suggest that T4 carcinoma can take full advantage of this mechanism.
Apoptosis is a physiologic process that tends to eliminate unnecessary cells or tumor cells from normal tissue [10]. Therefore, apoptosis plays a major role in the maintenance of tissue homeostasis. Failure of tumor cells to undergo apoptosis can result in their uncontrolled accumulation [10]. The AI in T4 carcinoma was lower than that in large T3 carcinoma, but not significantly. Although cell immortality of T4 carcinoma was not demonstrated in this study, further examination should be done concerning apoptosis in T4 carcinoma.
The density of TIL has received attention as a reflection of host-dependent immunologic status and as a useful prognostic indicator in some cancers including esophageal carcinoma [12, 13]. Most T4 and large T3 carcinomas show low TIL density. T4 carcinomas are associated with host defense activity as are large T3 carcinomas, but growth signals increase significantly in T4 carcinomas. It is difficult to grow single normal cells with an in vitro culture system [20]. Growth signals and cell immortality increase in single cells such as the p53 gene knockout cell and BC12 transgene cell, and these cells can easily grow in vitro [20]. Given these previous findings a few T4 carcinoma cells in which growth signals increase may possibly grow more easily than T2 or T3 carcinoma cells in the thoracic cavity after operative implantation or in the capillary bed of the distant organ after transport in the circulation.
In conclusion, T4 esophageal carcinoma shows a low expression of p27 and a high expression of cyclin A, which represents accelerated growth signals, and this carcinoma stage shows a low TIL density, which indicates a weakened host defense. The combination of p27 and cyclin A is a useful independent prognostic factor, and as a biological index is a useful measure of the progress of this carcinoma. TIL density is also a useful prognostic factor beyond conventional immunonutritional factors such as serum albumin or the total lymphocyte count.
|
References
|
|---|
-
Orringer M.B., Forastiere A.A., Perez-Tamayo C., Urba S., Takasugi B.J., Bromberg J. Chemotherapy and radiation therapy before transhiatal esophagectomy for esophageal carcinoma. Ann Thorac Surg 1990;49:348-355.[Abstract]
-
Hoff S.J., Stewart J.R., Sawyers J.L., et al. Preliminary results with neoadjuvant therapy in resection for esophageal carcinoma. Ann Thorac Surg 1993;56:282-287.[Abstract]
-
Ancona E., Ruol A., Castoro C., et al. First-line chemotherapy improves the resection rate and long-term survival of locally advanced (T4, any N, M0) squamous cell carcinoma of the thoracic esophagus. Ann Surg 1997;226:714-724.[Medline]
-
Levard H., Pouliquen X., Hay J.M., et al. 5-Fluorouracil and cisplatin as palliative treatment of advanced oesophageal squamous cell carcinoma. Eur J Surg 1998;164:849-857.[Medline]
-
Tabira Y., Okuma T., Kondo K., Kitamura N. Indications for three-field dissection followed by esophagectomy for advanced carcinoma of the thoracic esophagus. J Thorac Cardiovasc Surg 1999;117:239-245.[Abstract/Free Full Text]
-
Chen L.Q., Hu C.Y., Ghadirian P., Duranceau A. Early detection of esophageal squamous cell carcinoma and its effects on therapy. Dis Esophagus 1999;12:161-167.[Medline]
-
Rice T.W., Zuccaro G., Jr, Adelstein D.J., Rybicki L.A., Blackstone E.H., Goldblum J.R. Esophageal carcinoma. Ann Thorac Surg 1998;65:787-792.[Abstract/Free Full Text]
-
Motokura T., Arnold A. Cyclins and oncogenesis. Biochim Biophys Acta 1993;1155:68-78.
-
Elias J.M. Cell proliferation indexes. Biotechnol Histochem 1997;72:78-85.
-
Kerr J.F.R., Winterford C.M., Harmon B.V. Apoptosis. Cancer 1994;73:2013-2026.[Medline]
-
Chouaib S., Asselin-Paturel C., Mami-Chouaib F., Caignard A., Blay J.Y. The host-tumor immune conflict. Immunol Today 1997;18:493-497.[Medline]
-
Aaltomaa S., Lipponen P.K., Eskelinen M.J., et al. Lymphocyte infiltrates as a prognostic variable in female breast cancer. Eur J Cancer 1992;28A:859-864.
-
Ikeguchi M., Saito H., Katano K., Tsujitani S., Maeta M., Kaibara N. Correlation between the lymphocytic infiltration of tumors and the proliferative activity of cancer cells from surgically treated esophageal carcinoma. Oncology 1997;54:311-317.[Medline]
-
Ohashi Y., Sasano H., Yamaki H., et al. Cell cycle inhibitory protein p27 in esophageal squamous cell carcinoma. Anticancer Res 1999;19:1843-1848.[Medline]
-
Kinugasa S., Tachibana M., Hishikawa Y., et al. Prognostic significance of proliferating cell nuclear antigen (PCNA) in squamous cell carcinoma of the esophagus. Jpn J Clin Oncol 1996;26:405-410.[Abstract/Free Full Text]
-
Furihata M., Ishikawa T., Inoue A., et al. Determination of the prognostic significance of unscheduled cyclin A overexpression in patients with esophageal squamous cell carcinoma. Clin Cancer Res 1996;2:1781-1785.[Abstract]
-
Mineta Y., Miura K., Suzuki I., et al. Low p27 expression correlates with poor prognosis for patients with oral tongue squamous cell carcinoma. Cancer 1999;85:1011-1017.[Medline]
-
Fan G.K., Fujieda S., Sunaga H., et al. Expression of p27 is associated with progression and prognosis in laryngeal cancer. Laryngoscope 1999;109:815-820.[Medline]
-
Cooper S. A unifying model for the G1 period in prokaryotes and eukaryotes. Nature 1979;280:17-19.
-
Yasunaga M., Adachi S., Itoh N., Nishikawa S.I. Making the in-vitro model closer to actual B lymphopoiesis in the bone marrow. Semin Immunol 1995;7:186-195.
Accepted for publication June 26, 2000.
Top
Abstract
Introduction
