|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 1999 American Society for Clinical Oncology Prognostic Significance of Apoptotic Index in Completely Resected Non–Small-Cell Lung CancerFrom the Department of Thoracic Surgery, Faculty of Medicine, Kyoto University, Kyoto, Japan. Address reprint requests to Hiromi Wada, MD, Department of Thoracic Surgery, Faculty of Medicine, Kyoto University, Shogoin-kawahara-cho 53, Sakyo-ku, Kyoto, 606-8397, Japan; email wada{at}frontier.kyoto-m.ac.jp
PURPOSE: To evaluate the significance of apoptotic index (AI) as a prognostic factor after surgery for non–small-cell lung cancer (NSCLC). PATIENTS AND METHODS: A total of 236 patients who underwent surgery for previously untreated pathologic stage I to IIIa NSCLC between 1985 and 1990 were reviewed. AI was defined as the number of apoptotic cells, detected by terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate-biotin nick end-labeling, per 1,000 tumor cells. Proliferative index (PI) and aberrant p53 expression were also evaluated immunohistochemically.
RESULTS: The 5-year survival rate for the lowest-AI group (AI < 5.0) was 74.7%; those for the lower-AI group (5.0 CONCLUSION: AI proved to be an independent prognostic factor in NSCLC.
NON–SMALL-CELL LUNG cancer (NSCLC) is a malignant tumor with poor prognosis. To improve the prognosis, it is important to establish biologic markers, other than tumor, node, metastasis factors, that determine prognosis and response toward a particular treatment. Although many possible biologic markers, such as abnormality of p53 and c-myc, have been investigated, none of them has been established as a marker in the treatment of NSCLC.1 Apoptosis, a mode of cell death clearly distinct from necrosis, is a physiologic phenomenon that occurs spontaneously in the process of normal tissue growth.2 Apoptosis has been also shown to be closely associated with malignant tumors, and abnormality in the process of apoptosis promotes malignant transformation and leads to tumor proliferation. Apoptosis is induced by anticancer agents and irradiation, and abnormalities of several genes regulating apoptosis have been reported to alter sensitivity of tumor cells against chemotherapy and radiotherapy.3 Despite all these implications of relationships between apoptosis and cancer described above, the significance of apoptosis in cancer as a biologic marker, especially as a prognostic factor, has not been established. Many previous investigations conducted on this subject yielded contradictory results; some reports demonstrated that high apoptosis led to poor prognosis, some reported that high apoptosis led to good prognosis, and others demonstrated that apoptosis was not related to prognosis. Most of the previously reported results suggest that apoptosis is a form of cell death as a secondary result of active cell division and proliferation.4 For example, apoptosis is reported to occur more frequently in colorectal adenomas with severe dysplasia (containing cells proliferating more rapidly) compared with those with mild dysplasia.5 Also, apoptotic cells are observed more frequently in undifferentiated carcinomas (where cells are proliferating more actively) than in differentiated tumors. The higher tumor malignancy is related to the more active cell proliferation as well as the higher content of apoptotic cells, and the number of apoptotic cells increases with tumor size and progression of lymph node metastasis.6-9 Furthermore, an examination of apoptotic cell death in colorectal carcinomas with lymph node and/or liver metastases revealed that the fractions of both apoptotic cells and proliferative cells were significantly higher in metastatic foci than in primary lesions, implying that the observed increase in number of apoptotic cells reflects the higher activity of cell division in metastases.10 These observations have led to a conclusion that a higher content of apoptotic cells in malignant tissue indicates more rapid tumor-cell division and, consequently, results in the poorer prognosis.6,11,12,13,14 These results, however, are not consistent with the following observations of previous studies on apoptosis: fractions of apoptotic cells and proliferative cells were inversely correlated in esophageal carcinomas15; the content of apoptotic cells was high in slowly growing basal cell carcinomas16; examinations of apoptosis in gastric carcinoma17 and esophageal carcinoma18 showed that the fraction of apoptotic cells was higher in undifferentiated tumors than in differentiated ones; an investigation on proliferative and apoptotic fractions in non-Hodgkin's lymphomas identified a higher fraction of apoptotic cells in non–tumor region than in tumor regions19; and the fraction of apoptotic cells was higher in gastric mucosa dysplasias than in gastric carcinomas.20 Moreover, reports have shown that tumors with a high fraction of apoptotic cells tend to have a reduced risk of distant metastasis21 and a better prognosis22 because of increased cancer-cell death. At the same time, several reports demonstrated that the fraction of apoptotic cells did not correlate with tumor progression or malignancy23,24 or relate to prognosis.19 Thus, no reasonable interpretation of these contradictory observations on the role of apoptosis in the regulation of tumor-cell proliferation has been presented to date. It is particularly important, when discussing tumor progression, to consider balance of both cell death and cell proliferation because an argument based on only one of these factors may lead to an incorrect or confused conclusion. Nevertheless, attention has been focused in preceding investigations exclusively on whether the fraction of either apoptotic or proliferative cells is high or low in a particular malignant tissue. In the present study, we conducted a detailed investigation on the balance of apoptosis and cell proliferation in patients undergoing resection of NSCLC and clearly demonstrated that apoptosis has a biologic significance as well as a clinical significance as a prognostic factor. Apoptosis was detected by terminal deoxynucleotidyl transferase–mediated deoxyuridine triphosphate-biotin nick end-labeling (TUNEL),25 which detects not only cells with typical morphologic features of apoptosis but also cells in an earlier stage of apoptosis showing no apparent morphologic change. Cell proliferation was evaluated by immunohistochemical detection of proliferating cell nuclear antigen (PCNA),26,27 which is reported to be expressed in the cell nucleus during late G1 and S stages of the cell cycle. Aberrant expression of p53,28,29 a gene playing an important role in regulating apoptotic process, was also investigated in relation to apoptosis.
Clinical Characteristics of Patients A total of 237 consecutive patients with pathologic (p)-stage I to IIIa NSCLC, who underwent complete tumor resection and mediastinal lymph node dissection without any preoperative therapy at the Department of Thoracic Surgery, Kyoto University between January 1, 1985, and December 31, 1990, were reviewed (Table 1). Complete tumor resection was considered achieved when no microscopic cancer cells were identified in either the margin of resection of the tumor or the highest mediastinal lymph nodes.30 P-stage was re-evaluated and determined by the present tumor, node, metastases classification as revised in 1997.31 Histologic type and cell differentiation were determined using the classification by the World Health Organization.32 One patient was excluded from the study because of operation-related death, and thus, a final total of 236 patients were evaluated. For all these patients, the inpatient medical records, chest x-ray films, whole-body computed tomography films, bone and gallium scanning data, and records of surgery were reviewed without knowledge of the results of TUNEL or immunohistochemical staining. Intraoperative therapy was not performed on any patient. As postoperative adjuvant therapy, cisplatin-based chemotherapy, radiation, and oral administration of tegafur (a fluorouracil derivative drug) were prescribed for 55, 35, and 58 patients, respectively.33 Follow-up of the postoperative clinical course was conducted by outpatient medical records and by inquiries by telephone or letter. The follow-up survey was successfully completed for 100% of patients for 5 years after surgery. The day of thoracotomy was considered the starting day for counting postoperative survival days.
Tissue Preparation All tumor specimens were immediately fixed in 10% (v/v) formalin and then embedded in paraffin. Serial 4-µm sections were prepared from each sample and served for routine hematoxylin and eosin (HE) staining, TUNEL staining to detect apoptotic cells, and immunohistochemical staining to detect PCNA expression and p53 aberrant expression. For the TUNEL staining and immunohistochemical staining, the sections were dewaxed by heating for 30 minutes at 60°C followed by washing in xylene (Nakalai Tesque, Kyoto, Japan) for 5 minutes three times each. The dewaxed sections were dehydrated by immersion in 100%, 100%, 90%, 80%, 70%, and 50% ethanol solutions (Nakalai Tesque) for 3 minutes each. After they were washed in phosphate-buffered saline (PBS), the sections were digested with 20 µg/mL proteinase K (Boehringer Manheim, Manheim, Germany) for 20 minutes at 25°C for the TUNEL staining; for immunohistochemical detection of PCNA and p53 expression, the sections were heated in a microwave oven for 5 minutes three times each to retrieve their antigeneities.34 Endogenous peroxidase was inactivated by incubating the sections with 0.03% H2O2 in methanol (Nakalai Tesque) for 30 minutes at 25°C, and then the sections were washed in PBS. To reduce nonspecific labeling, the sections were incubated with normal calf serum (DAKO Japan, Kyoto, Japan).
Detection of Apoptotic Cells The specificity of the TUNEL staining of apoptotic cells was confirmed by making negative and positive control slides at every staining. Sections incubated with the TUNEL reaction mixture without TdT were used as negative control slides. Sections treated with 0.7 mg/mL of DNase I (Stratagene, La Jolla, CA) for 10 minutes at 25°C before the TUNEL reaction were used as positive control slides. Apoptotic cells were determined with careful observation of TUNEL-stained sections and serial HE-stained sections because some necrotic cells could be also TUNEL-positive. If TUNEL-positive stained cells did represent histologic features of necrosis in HE-stained sections, they were not considered to be apoptotic cells. In each case, a total of 10,000 tumor cells, 1,000 tumor cells each in 10 different fields, were evaluated at high magnification (x400) by two authors independently (F.T., Y.O.), without knowledge of patients' characteristics or clinical data. A different evaluation of apoptotic cells was made (F.T., Y.O.) in 11 patients (4.7%). The field was re-evaluated until the evaluation coincided. The apoptotic index (AI) was defined as the number of apoptotic cells per 1,000 tumor cells.
Immunohistochemistry
Statistical Methods
Fraction of Apoptotic Cells in NSCLC The mean AI ± SE calculated for all 236 patients studied was 18.8 ± 1.4, and the median AI was 11. AI was significantly lower in older patients than in younger patients; no significant correlation was observed between AI and sex, performance status (PS), histologic type, or p-stage (Table 1).
In previously reported studies on fraction of apoptotic cells, patients were simply divided into two groups, higher-AI and lower-AI, and the clinical characteristics and the prognoses were compared between the groups. In the present study, to investigate the clinical significance of AI in more detail, the distribution of AI was examined first. A patient group with extremely low AI values (below 5.0, the 25th percentile value of AI) was clearly distinguishable. At the same time, two patient groups with intermediate AI values and a third patient group with the highest AI values (above 25.0, the 75th percentile value of AI) were also identified. Thus, all patients were divided into the following four groups, on the basis of AI, using the 25th, 50th, and 75th percentile values (5.0, 11.0, and 25.0, respectively): group A, the lowest-AI group (AI < 5.0, n = 59); group B, the lower-AI group (5.0
AI and Prognosis After Resection
The results presented above were obtained for all patients studied who had p-stage I to IIIa NSCLC. In an attempt to investigate prognosis in a more homogeneous patient group, only p-stage I patient prognosis was examined because the majority of the patients studied had p-stage I disease. The results were similar with those obtained for all patients. The 5-year survival rate for group A (n = 36) was 87.9%, demonstrating good prognosis. In contrast, 5-year survival rates for group B (n = 29) and group C (n = 35) were 59.4% and 68.5%, respectively, which demonstrated that the prognosis of these moderate AI groups was significantly worse compared with group A. Interestingly, the 5-year survival rate for group D (n = 38) was as high as 89.2%, demonstrating that patients with the highest AI (25.0
AI and Proliferative Fraction
PIs for the four AI groups are listed in Table 3. For all patients with p-stage I to IIIa disease, group A had the lowest PI of the four groups (Fig 3), which showed that overall tumor growth was slow in group A, the lowest-AI group. Consequently, group A had a good prognosis. On the other hand, PIs for group B and group C were 48.0% and 54.3%, respectively, significantly higher than the PI of group A (P = .002 for group B and P < .001 for group C). These results indicate that, in group B and group C, PI was high and the number of apoptotic cells increased as a result of increased cell proliferation. Under such conditions, increased cancer-cell proliferation would lead to poor prognosis. In group D, the highest AI group, high activity of cell proliferation (PI = 50.7%) might be overcome by predominating cell death because of apoptosis, resulting in retarded tumor-cell proliferation and good prognosis. Analysis stratified by p-stage also demonstrated that the PI in group A the was lowest among the four AI groups in any p-stage (Table 3).
AI and Tumor Differentiation
AI and p53 Status
Multivariate Analysis of Prognostic Factors
In this article, the biologic and clinical significance of AI in NSCLC was clearly demonstrated for the first time. Detailed examination of AI in lung cancer tissue identified two borderline values, five apoptotic cells per 1,000 tumor cells and 25 apoptotic cells per 1,000 tumor cells, that determined postoperative prognosis. In group A, the lowest-AI group (AI < 5.0), prognosis after surgical resection proved to be relatively good. In groups B and C, the moderate AI groups (5.0  AI < 25.0), prognosis was poor. Surprisingly, however, prognosis was excellent in group D, the highest-AI group (25.0 AI). These apparently contradictory results are reasonably interpreted by taking cell proliferation rate into consideration. In group A, the number of tumor cells killed by the mechanism of apoptosis was small, most presumably because the number of proliferative tumor cells was small; slow tumor proliferation have led to good prognosis in group A. In groups B and C, tumor cells divided actively and apoptotic cells increased as a consequence of the accelerated cell proliferation. Under such conditions, cell proliferation might have been predominant on the whole rather than cell death, and accelerated tumor-cell proliferation would have resulted in poor prognosis. In group D, apoptotic tumor-cell death might have overcome tumor-cell proliferation to retard tumor growth and improve prognosis. The results and interpretation presented above provide reasonable explanation for previously reported contradictory observations on AI in tumor tissue. In a well-differentiated tumor tissue, such as the keratinizing region of SCC of the esophagus, apoptosis was induced more frequently than in the nonkeratinizing region,15 possibly because the mechanism of apoptosis against cell proliferation occurred normally in a well-differentiated and keratinizing region. On the other hand, in one report, the number of apoptotic cells was significantly lower in differentiated gastric carcinomas than in undifferentiated tumors,7 possibly because the activity of cell proliferation was relatively higher in less-differentiated tumor tissue than in well-differentiated tissue, and, as a result, the active cell proliferation led to an increase in the number of apoptotic cells. In the present study, we have also demonstrated that AI in well-differentiated tumor tends to be either low or extremely high. In these well-differentiated tumors, the cell proliferation rate may be basically slow and the number of apoptotic cells may be consequently small. If the cell proliferation rate increases to exceed a certain limit, however, the mechanism of apoptotic regulation may properly operate to kill overproliferating tumor cells and suppress acceleration of tumor-cell proliferation. Contradictory data have also been accumulated on the relationship between apoptosis and prognosis. Compared with patients with an extremely low number of apoptotic cells, patients with a relatively higher number of apoptotic cells had worse prognoses because of increased cancer-cell proliferation. On the other hand, a good prognosis was observed when elevated apoptotic activity overcame cell proliferation. This apparent contradiction may be derived from the fact that, in previously reported studies, the patients studied were simply classified into two groups based solely on whether the number of apoptotic cells was higher or lower. Our results, obtained by dividing the patients into four groups based on the distribution of AI, demonstrated that well-differentiated tumors belonged to either group A (where cell proliferation is slow) or group D (where apoptosis is capable of compensating accelerated cell proliferation) and that well-differentiated tumors had a good prognosis as a result of well-balanced cell death and cell proliferation in either group A or group D. Clinical significance of apoptosis in NSCLC has not been established. Komaki et al24 reported that AI positively correlated with PI but showed no correlation between differentiation, size, or tumor prognosis. On the other hand, Tormanen et al14 reported that enhanced apoptosis resulted in shortened survival. As discussed above for cancer in other organs, an examination of the influence of apoptosis on lung cancer prognosis, based solely on whether the number of apoptotic cells was higher or lower, would yield contradictory results. In the present study, it was clearly demonstrated that apoptosis, viewed in relation to PI, served as a factor to determine postoperative survival in NSCLC. In this study, the TUNEL method was used to detect apoptotic cells. With routinely used HE staining, apoptotic cells can be identified morphologically. However, with HE staining, some apoptotic cells at earlier stages exhibiting no morphologic changes characteristic of apoptosis cannot be detected, or it may be morphologically difficult to identify apoptotic cells. The TUNEL method detects fragmented DNA in apoptotic cells25 and yields reliable results even when paraffin-embedded sections are used.37 Using this technique, apoptotic cells at earlier stages, with no apparent morphologic features of apoptosis, can be histologically detected,17 and quantitative determination of apoptotic cells can be achieved with high accuracy and sensitivity.20 Because the TUNEL method detects fragmentation of DNA through causes other than apoptosis,38 necrotic cells give a positive result.39 Therefore, we closely compared results obtained with the routine HE staining and the TUNEL method for exclusion of necrotic cells and specific identification of apoptotic cells. The p53 gene regulates the apoptotic process as well as the cell cycle.28,29 When genome is injured or damaged in some cells, wild-type p53 prevents cell division by arresting cells at a particular stage of the cell cycle and induces apoptosis in these cells to kill them and prevent generation and proliferation of transformed cells.40 Mutation of p53 has been detected in many malignant tumors, including NSCLC. Mutated p53 fails to control proliferation of transformed cells and, consequently, allows tumor-cell proliferation. Several investigators have observed no correlation between apoptosis and abnormality of p53,10,15,18 which has been ascribed to inaccuracy in immunohistochemical detection of p53 abnormalities derived from false-positive and false-negative reactions41 or existence of a p53-independent pathway for apoptosis induction.42 In colorectal43 and gastric cancer,44 on the other hand, aberrant p53 expression has been reported to be correlated with abnormality of p53 gene, and significant suppression of apoptosis has been observed in cases with a mutated p53 gene. Although we have observed no correlation between apoptotic fraction and aberrant p53 expression in the present study, further investigation that takes into consideration mutation in the p53 gene itself is needed on this subject. Although apoptosis may be induced spontaneously in previously untreated tumor tissues, as shown in the present study, it has been proved both experimentally and clinically that apoptosis is also induced by anticancer agents or irradiation. Moreover, mutation of p53 has been shown to alter the sensitivity of tumor cells toward anticancer agents or irradiation. This study demonstrates the clinical significance of apoptosis as a prognostic factor after surgical resection. In future investigations, the clinical significance of AI as a factor to predict the therapeutic effect of chemotherapy will be examined.
We thank Tomoko Yamada for excellent preparation of histologic sections.
1. American Society of Clinical Oncology:Clinical practical guidelines for the treatment of unresectable non–small-cell lung cancer. J Clin Oncol 15:2996-3018, 1997[Abstract] 2. Kerr JF, Wyllie AH, Currie AR: Apoptosis: A basic biological phenomenon with wide-ranging implication in tissue kinetics. Br J Cancer 26:239-257, 1972[Medline] 3. Kerr JFR, Winterford CM, Harmon BV: Apoptosis: Its significance in cancer and cancer therapy. Cancer 73:2013-2026, 1994[Medline] 4. Arends MJ, McGregor AH, Wyllie AH: Apoptosis is inversely related to necrosis and determines net growth in tumors bearing constitutionally expressed myc, ras, and HPV oncogenes. Am J Pathol 144:1045-1057, 1994[Abstract] 5. Arai T, Kino I: Role of apoptosis in modulation of the growth of human colorectal tubular and villous adenomas. J Pathol 176:37-44, 1995[Medline] 6. Vesalainen S, Lipponen P, Talja M, et al: Histological grade, perineural infiltration, tumor-infiltrating lymphocytes and apoptosis as determinants of long-term prognosis in prostatic adenocarcinoma. Eur J Cancer 30A:1797-1803, 1994 7. Saegusa M, Takano Y, Wakabayashi T, et al: Apoptosis in gastric carcinomas and its association with cell proliferation and differentiation. Jpn J Cancer Res 85:939-945, 1994[Medline] 8. Aihara M, Truong LD, Dunn JK, et al: Frequency of apoptotic bodies positively correlates with Gleason grade in prostate cancer. Hum Pathol 25:797-801, 1994[Medline] 9. Gaffney EF: The extent of apoptosis in different types of high grade prostatic carcinoma. Histopathology 25:269-273, 1994[Medline] 10. Tanabe S, Ishida M, Kasagi N, et al: Apoptosis occurs more frequently in primary lesions of human colorectal carcinomas: Analysis by terminal-deoxynucleotidyl-transferase-mediated dUTP-biotin neck end labeling. Int J Cancer 65:173-177, 1996[Medline] 11. Leoncini L, Del Vecchio MT, Megha T, et al: Correlation between apoptotic and proliferative indices in malignant non-Hodgkin's lymphomas. Am J Pathol 142:755-763, 1993[Abstract] 12. Levine EL, Davidson SE, Roberts SA, et al: Apoptosis as a predictor of response to radiotherapy in cervical carcinoma. Lancet 344:472, 1994[Medline] 13. Aihara M, Scardino PT, Truong LD, et al: The frequency of apoptosis correlates with the prognosis of Gleason grade 3 adenocarcinoma of the prostate. Cancer 75:522-529, 1995[Medline]
14.
Tormanen U, Eerola AK, Rainio P, et al: Enhanced apoptosis predicts shortened survival in non-small cell lung carcinoma. Cancer Res 55:5595-5602, 1995 15. Ohbu M, Saegusa M, Okayasu I: Apoptosis and cellular proliferation in oesophageal squamous cell carcinomas: Between keratinizing and nonkeratinizaing types. Virchows Arch 427:271-276, 1995[Medline] 16. Kerr JFR, Searle JA: Suggested explanation for the paradoxically slow growth rate of basal-cell carcinomas that contain numerous mitotic figures. J Pathol 107:41-44, 1972[Medline] 17. Kasagi N, Gomyo Y, Shirai H, et al: Apoptotic cell death in human gastric carcinoma: Analysis by terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick and neck labeling. Jpn J Cancer Res 85:939-945, 1994
18.
Hamada M, Naomoto Y, Fujiwara T, et al: Squamous apoptotic induction in esophageal squamous cell carcinomas expressing extensive p53 protein. Jpn J Clin Oncol 26:398-404, 1996 19. Czadar M, Mazur J, Petterson M, et al: Prognostic significance of proliferative and apoptotic fractions in low grade follicle center cell-derived non-Hodgkin's lymphomas. Cancer 77:1180-1188, 1996[Medline] 20. Ishida M, Gomyo Y, Tatebe S, et al: Apoptosis in human gastric mucosa, chronic gastritis, dysplasia and carcinoma: Analysis by terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end labeling. Virchows Arch 428:229-235, 1996[Medline] 21. Radinsky R, Fidler IJ, Prince JE, et al: Terminal differentiation and apoptosis in experimental lung metastasis of human osteogenic sarcoma cells by wild type p53. Oncogene 9:1877-1883, 1994[Medline] 22. Wheeler JA, Stephans LC, Tornos C, et al: ASTRO research fellowship: Apoptosis as a predictor of tumor response to radiation in stage IB cervical carcinoma. Int J Radiat Oncol Biol Phys 32:1487-1493, 1995[Medline] 23. Brown C, Sauvageot J, Kahane H, et al: Cell proliferation and apoptosis in prostate cancer: Correlation with pathologic stage? Mod Pathol 9:205-209, 1996[Medline] 24. Komaki R, Fujii T, Perkins P, et al: Apoptosis and mitosis as prognostic factors in pathologically staged N1 nonsmall cell lung cancer. Int J Radiat Oncol Biol Phys 36:601-605, 1996[Medline]
25.
Gavrieli Y, Sherman Y, Ben-Sasson SA: Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J Cell Biol 119:493-501, 1992 26. Bravo R, Frank R, Blundell PA, et al: Cyclin/PCNA is the auxiliary protein of DNA polymerase delta. Nature 326:515-517, 1987[Medline] 27. Ishida M, Gomyo Y, Ohfuji S, et al: Evidence that expression of a mutated p53 gene attenuates apoptotic cell death in human gastric interstitial-type carcinomas in vivo. Jpn J Cancer Res 88:468-475, 1997[Medline] 28. Levine AJ, Monmand J, Finlay CA: The p53 tumor supressor gene. Nature 351:453-456, 1991[Medline] 29. Vogelstein B, Kinzler KW: p53 function and dysfunction. Cell 70:523-526, 1992[Medline]
30.
Wada H, Tanaka F, Yanagihara K, et al: Time trends and survival after surgery for primary lung cancer during 1976-1990. J Thorac Cardiovasc Surg 112:349-345, 1996 31. Naruke T, Tsuchiya R, Kondo H, et al: Implication of staging in lung cancer. Chest 112:242s-248s, 1997 32. World Health Organization:The World Health Organization histological typing of lung tumors: Second edition. Am J Clin Pathol 77:123-136, 1982[Medline] 33. Tanaka F, Yanagihara K, Ohtake Y, et al: p53 status predicts the efficacy of postoperative oral administration of tegafur for completely resected non-small cell lung cancer. Jpn J Cancer Res 90:432-438, 1999[Medline] 34. Campani D, Cecchetti D, Bevilacqua G: Immunocytochemical p53 detection by microwave oven heating of routinely-fixed paraffin section. J Pathol 171:151-152, 1993[Medline] 35. Sgons R, Wick G: Methods for the detection of apoptosis. Int Arch Allergy Immunol 105:327-332, 1994[Medline]
36.
Tanaka F, Miyahara R, Ohtake Y, et al: Lewis Y antigen expression and postoperative survival in nonsmall cell lung cancer. Ann Thorac Surg 66:1745-1750, 1998 37. Wijsman JH, Jonker RR, Keijzer R, et al: A new method to detect apoptosis in paraffin sections: In situ end-labeling of fragmented DNA. J Histochem Cytochem 41:7-12, 1993[Abstract] 38. Asari B, Coates PJ, Greenstein BD, et al: In situ end-labeling detects DNA strand breaks in apoptosis and other physiological states. J Pathol 170:1-8, 1993[Medline] 39. Gold R, Schmied M, Giegerich G, et al: Differentiation between cellular apoptosis and necrosis by the combination use of in situ tailing and nick translation techniques. Lab Invest 71:219-225, 1994[Medline] 40. Lane DP: p53: Guardian of the genome. Nature 358:15-16, 1992[Medline] 41. Carbone DP, Mitsudori T, Chiba I, et al: p53 immunostaining positivity is associated with reduced survival and imperfectly correlated with gene mutations in resected non-small cell lung cancer: A preliminary report of LCSG 871. Chest 106:377s-381s, 1994 42. Clarke AR, Purdie CA, Harrison DJ, et al: Thymocyte apoptosis induced by p53-dependent and independent pathways. Nature 362:849-852, 1993[Medline] 43. Kobayashi M, Watanabe H, Ajioka Y, et al: Correlation of p53 protein expression with apoptotic incidence in colorectal neoplasm. Virchows Arch 427:27-32, 1995[Medline] 44. Ishida M, Gomyo Y, Ohfuji S, et al: Evidence that expression of a mutated p53 gene attenuates apoptotic cell death in human gastric interstitial-type carcinomas in vivo. Jpn J Cancer Res 88:468-475, 1997 Submitted February 16, 1999; accepted May 21, 1999. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
AI < 11.0) and the higher-AI group (11.0 
2 test was used to compare counts and analyze trends in counts. Continuous data were compared using Student's t test, if the distribution of samples was normal, or the Mann-Whitney U test, if the sample distribution was asymmetrical. The postoperative survival rate was analyzed by the Kaplan-Meier method, and differences in survival rates were assessed by the log-rank test. Multivariate analysis of prognostic factors was performed using Cox's regression model. Differences were considered significant when P < .05. All statistical manipulations were performed using the SPSS for Windows software system (SPSS Inc, Chicago, IL). 
, mean ± SE) were determined by TUNEL staining and by immunohistochemistry using monoclonal antibody against PCNA.
), 0 
), 5 
