- Email: [email protected]
Immunohistochemical analysis of proliferating cell nuclear antigen expression in human neuroblastoma
Immunohistochemical Analysis of Proliferating Cell Nuclear Antigen Expression in Human Neuroblastoma By Takaharu Oue, Masahiro Fukuzawa, Shinkichi Kamata, and Akira Okada Osaka, Japan • Proliferative cell nuclear antigen (PCNA) is a cell-cyclerelated nuclear protein that is maximally elevated in late G1 and S phase of proliferating cells. In this study, PCNA was identified immunohistochemically using paraffin section in 67 human neuroblastomas. Percentage of the PCNA-positive nuclei (PCNA index: PCI) ranged from 0% to 75%. There were significant relations between the PCNA expression and mitotic karyorrhexis index (MKI}, histological classification, cell concentration, tumor weight, clinical stage, local invasion, lymph node metastasis, liver metastasis, or DNA ploidy. PCI was significantly low in patients who received aggressive chemotherapy before surgery. Patients with PCI higher than 30% showed a worse survival rate compared with those with PCI lower than 10% (P < .01). High PCI significantly related with poor survival, so that PCI may be an independent prognostic factor in neuroblastoma. Although further studies are required, PCNA immunostaining may be useful for assessing proliferating activity and for providing prognostic information in human neuroblastoma. Copyright © 1995 by W.B, Saunders Company INDEX WORDS: Proliferating cell nuclear antigen, neuroblastoma, immunohistochemistry.
NFORMATION on cellular proliferation would be useful for predicting metastatic potential, recurrence, and overall prognosis in various human neoplasms.I, 2 Proliferating cell nuclear antigen (PCNA) is a cell-cycle-related nuclear protein that is maximally elevated in late G1 and S phase. 3-s It has recently been identified as the polymerase 0 accessory protein; therefore its level correlates directly with the rate of cellular proliferation and DNA synthesis. 6-8 A monoclonal antibody, PC10, has been shown to recognize PCNA in formalin-fixed, paraffin-embedded tissues. T M Neuroblastoma is the most common pediatric solid neoplasm. Although some biological variables such as N-myc gene amplification or DNA ploidy are reported to have relations with prognosis of this tumor, the relation between proliferating activity and prognosis is unknown. The purpose of this study is to analyze PCNA expression in human neuroblastoma using
I
From the Department of Pediatric Surgery, Osaka University Medical School, Osaka, Japan. Address reprint requests to Takaharu Oue, Department of Pediatric Surgery, Osaka University Medical School, 2-2 Yamadaoka, Suita City, Osaka 565, Japan. Copyright © 1995 by W.B. Saunders Company 0022-3468/95/3004-0004503. O0/ 0 528
immunohistochemical staining with PC10, and to assess whether it represents proliferating activity of neuroblastoma cells and to provide prognostic information. MATERIALS AND METHODS
Patient Profiles Samples of neuroblastoma were available from 67 patients who were operated on at Osaka University Hospital from 1970 to 1993. They were 36 boys and 31 girls, with a mean age of 2 years 2 months (from 14 days to 12 years of age). Eighteen of the patients were detected through mass screening. Histologically, 53 tumors were neuroblastoma and 14 were ganglioneuroblastoma. According to Evans' staging system, 13 patients were of stage I, 5 were of II, 18 were of III, 28 were of IV, and 3 were of IVs. Forty patients received chemotherapy before surgery.
Method Extirpated samples were immediately fixed with 10% formalin solution and processed with standard procedures for paraffin embedding. Conventional histological sections with 3 txm of thickness were cut and mounted on poly-L-lysine-coated glass slides. Immunostaining was performed with standard streptavidin-biotin immunoperoxidase method (LSAB kit; Dako Japan Corp; Kyoto, Japan). Sections were dewaxed in xylene, rehydrated through alcohol, and immersed in 3% hydrogen peroxide in methanol for 30 minutes to block endogenous peroxidase activity, and normal goat serum was applied for 10 minutes to reduce nonspecific antibody binding. The monoclonal antibody PC10, a murine IgG, was obtained from Dako corporation, which was used at a dilution of 20 times with 2 hours' incubation at room temperature. Biotinylated goat anti-mouse antibody was used as the linker molecule for 20 minutes, and sections were incubated in streptavidin-horseradish peroxidase complex for 10 minutes. Diamino-benzidine-hydrogen peroxidase was used to detect the presence of the peroxidase, and a light Mayer's hematoxylin counterstain was applied. To ensure consistency of PCNA staining, a known positive control was included in each round, and negative control was also included in which the primary antibody was replaced by phosphate-buffered saline (PBS).
Assessment of PCNA Staining To negate the tumor heterogeneity, at least five high-power (x 400) fields were randomly selected for cell counting. A minimum of 1000 tumor cells were counted in each case. Cells were considered positive for PCNA only when definite reddish-brown staining of the nucleus was identified. The PC10 labeling index, PCI, was defined as the percentage of positive tumor cell nuclei in the following way: Number of cells with positive staining PCI = Total number of tumor cells counted x 100 (%)
A s s e s s e d Variables Examinations were performed to determine whether there were significant relations between PCNA expression and various patho-
Journalof PediatricSurgery,Vol 30, No 4 (April), 1995:pp 528-532
PCNA IMMUNOSTAINING IN HUMAN NEUROBLASTOMA
logical and clinical variables such as mitotic karyorrhexis index, cell concentration, tumor weight and histological classification, age, clinical stage, local invasion, metastasis, and DNA ploidy. Mitotic karyorrhexis index (MKI) defined in Shimada's classification, 12 was determined by counting the number of mitotic figures and karyorrhexis in 5000 tumor cells. To analyze DNA ploidy, flow cytometry was performed using paraffin blocks as described by Hedley et al. 13 The groups were compared using unpaired t test. For multifactorial analysis, analysis of variance (ANOVA) test was applied. Survival curve according to PCNA expression was calculated using the Kaplan-Meyer method, and the Generalized Wilcoxon test was applied to compare the survival rates.
RESULTS
PCNA Staining Figure 1 shows the typical nuclear staining of samples. Positive staining of PCNA was confined to the nucleus of proliferating cells. PCI varied within a range of 0% to 75%.
Effect of Chemotherapy During these 10 years, we have performed aggressive chemotherapy before surgery using cisplatin and doxorubicin for stage III and IV neuroblastoma patients. To access the effect of chemotherapy in this group, PCNA expression was compared between the patients who received chemotherapy and those who did not (Fig 2). PCI was significantly low in patients who received aggressive chemotherapy before surgery. To exclude the effect of chemotherapy, patients who received chemotherapy before surgery were excluded and 40 patients were analyzed for the following study.
Relation Between Pathological Variables and PCI PCI shows significant positive correlation with mitotic karyorrhexis index (Fig 3), cell concentration
529
PCI 80 %70
0
P<0.05 0
6050 40
ol
30 20 10
I
' -I-
chemotherapy Fig 2. Relation between chemotherapy and PCI. To access the effect of chemotherapy in this group, PCNA expression was compared between the patients who received chemotherapy and those who did not in the stage III and IV patients during these 10 years. PCI was significantly low in the patients who received aggressive chemotherapy before surgery. (O), dead; (©), alive.
(Fig 4), and tumor weight (Fig 5). Histologically, PCI in ganglioneuroblastoma is significantly lower than that in neuroblastoma (Table 1). PCI is low when differentiated cells appeared (Fig 6).
Relation Between Clinical Variables and PCI
Fig 1. PCNA-positive staining in stage IV neuroblastoma. Positive staining of PCNA was confined to the nucleus of proliferating cells.
PCI was significantly high in the cases of stage III and IV in comparison with the cases of stage I, II and IV. Tumors invading to other organs or with metastatic lesions in lymph nodes or liver show an increasing percentage of PCNA-positive cells. PCI was low in the cases under 1 year of age but not significant. PCI was significantly low in aneuploid tumors compared with diploid (Table 1).
530
OUE ET AL
2000
MKI
g
400
y=4.882x+4.484, r= 0.925 /
O
O ~
y =13.86 x - 2.58 r = 0.591
1500
300 1000 200
500
100 PCI
0 0
20
40
60
0
~ 0
% 80
PCl
20
40
60 %
Fig 3. Relationship between PCI and MKI. There is significant correlation between PCI and MKI.
Fig 5, Relationship between PCI and tumor weight, There is significant correlation between PCI and tumor weight.
Prognosis
recurrence, and overall prognosis. 12 PCNA is a 36 kDa nuclear protein, and its level is maximally elevated in late G1 and S phase and expression is closely linked to the cell cycle.3-5 It has recently been identified as the polymerase 0 accessory protein, and its level therefore correlates directly with the rate of cellular proliferation and DNA synthesis. 6-8 A monoclonal antibody, PC10, has been shown to recognize PCNA in formalin-fixed, paraffin-embedded tissues. 911 Current methods of assessing the proliferation and growth of tumors from slide-based materials with monoclonal antibodies such as Ki-67 were limited because they require cryostat sections of freshfrozen material. 1,2J4 The major advantage of staining with PC10 is its detectability in formalin-fixed paraffinembedded tissues, which enables us to perform retro-
Figure 7 shows survival cUrves that show the relationship between prognosis and PCI. Patients with PCI higher than 30% have a worse survival rate compared with that of patients whose PCI is lower than 10% (P < .01). These results suggest that a high PCI significantly related with poor survival and that PCI may be an independent prognostic factor in neuroblastoma. DISCUSSION Information on cell kinetics would be useful in understanding tumor behavior. In a variety of malignant neoplasms, correlations have been noted between proliferative activity and metastatic potential,
cell density 2.0 . . . . . . . .
rate of differentiated cells 30 . . . . . % o 25 o
' ....
/100~tm2 1.5
OOo
O
20
o e
1.0
~1~
15
°.4o°
10
0.5 0
0
O
+ °y=O.O2x r = 0.667
l 0
20
40
5
0.548
o
PCI 60 %
80
Fig 4. Relationship between PCI and cell concentration. There is significant correlation between PCI and cell density,
0
o
o o~ig ® ~ o
0
20
o
40
o
.
PCl
6O %
Fig 6. Relationship between cell differentiation and PCI. PC! is low in the tumor in which the rate of differentiated cells is higher than 5%.
PCNA IMMUNOSTAINiNG IN HUMAN NEUROBLASTOMA
531
Table 1. Relation Between Prognostic Variables and PCI
PCI Variables
Age < 1 yr >1 yr Stage I, II, IV III, IV Histology NB GNB Local invasion C1 C2 C3 Lymph node metastasis
n
(mean +- SD)
Probability
26 14
11.6 -+ 10.0 21.6 -+ 6.7
.078
19 21
7.4 -+ 9.9 22.1 -+ 19.5
.0051
32 7
18.4 - 17.6 0.35 -+ 0.41
.011
14
5.6 -+ 7,4 8.8 _+ 13.4 25,5 -+ 19.1
.05 (Cl versus C3)
7 18
n0, nl
24 13
8.5 -+ 10.4 27.0 _+ 22.1
.0014
n2, n 3
26 6
15.2 _+ 15.1 33.7 _+ 20.8
.017
10 17
30.5 -+ 25.0 14.1 -+ 9.7
.023
Liver metastasis + DNA ploidy diploid aneuploid
Abbreviations: NB, neuroblastoma; GNB, ganglioneuroblastoma; C~, localized to the capsule; C2, invasion to near organs; C3, invasion to opposite site; no, no metastasis; nl, local metastasis; n2, n3, distant metastasis.
spective and morphological evaluation on cellular proliferation. On the contrary, there are some problems in assessing proliferating activity using paraffin blocks. A partial loss of antigenicity of PCNA after formalin fixation cannot be excluded. Matsuno and Mukai 15showed that PCNA immunoreactivity is markedly reduced over 48 hours' formalin fixation, so materials should not be fixed longer than 48 hours. In our institute, all pathological materials were embedded and stained in our own pathological laboratory within 48 hours. *PCI<10 n=18
100 %
0,PO,
80
0
60 40
*30<_--PCI n=5 20 0
* P<0.01
1
2
4
year
Fig 7. Survival curve according to PCNA expression. Patients with PCI higher than 30% have a worse survival rate compared with these o PCI (P < ,01), with a lower than 10Vo
Recent reports suggest that this antibody may be useful in leukemias, lymphomas, and certain adult tumors as an independent prognostic variable,26-2° but no pediatric solid tumors have been studied in detail. In this study, we analyzed PCNA expression in human neuroblastoma using immunohistochemical techniques and assessed whether it represented proliferating activity and provided prognostic information. In order to assess the proliferating activity, Shimada et al defined the MKI, which is the number of mitotic figures and karyorrhexis in 5000 tumor cells, and applied it in their histological classification of neuroblastomas. 12 According to this classification, tumors with high MKI were classified as "unfavorable histology." In our study, PCI showed significant positive correlation with MKI. This result may indicate that PC! is considered to be an appropriate index of proliferation activity in neuroblastoma. PCI also showed significant positive correlation with tumor weight and clinical stage. Tumors with high proliferating activity may grow rapidly, so that tumor size may become larger and clinical stage may be advanced at the time of diagnosis. PCI was high in the tumors with local invasion and in the tumors with metastatic lesions in lymph nodes or liver. These results may indicate that proliferating activity plays an important role in both invasion and metastasis in neuroblastomas. Histologically, PCI in ganglioneuroblastoma was significantly lower than that in neuroblastoma. PCI was low in the tumors in which the rate of differentiated cells was higher than 5% (Fig 6). From these findings, differentiated cells are considered to have very low proliferating activity. Recent studies of DNA ploidy in neuroblastoma show that prognosis of the patients with aneuploid tumor is significantly good compared with the patients with diploid tumor. 21-23 In our study, PCI was significantly low in aneuploid tumors, which are considered to have poor proliferating activity or good prognosis in neuroblastoma. Woods et a119 showed that there is a significant relationship between PCI and S + G2M phase fraction as measured by flow cytometric analysis in gastrointestinal lymphomas. To the contrary, Jain et a124 showed that no significant correlation was observed in gastric carcinoma. We are investigating the relationship between PCI and S + G2M phase fraction in neuroblastoma, but sufficient correlation is not available (data not shown). One of the reasons is that it is difficult to recognize the S + G2M phase fraction in aneuploid tumors.
532
OUE ET AL
During these 10 years, we have performed aggressive chemotherapy using cisplatin and doxorubicin for advanced neuroblastoma patients. To access the effect of chemotherapy on the proliferative activity, PCNA expression was analyzed in the stage III and IV patients who had undergone surgery since 1980. The results show that proliferating activity was reduced by chemotherapy. Among chemotherapy groups, the patients whose tumor still showed a high percentage of PCI had a poor prognosis (Fig 1). These findings show that PCI may be used to access the effectiveness of chemotherapy and to predict the prognosis of a patient who received preoperative chemotherapy. Patients with a PCI higher than 30% have a worse survival rate compared with those who have a PCI
lower than 10%. These results suggest that a high PCI is significantly related to poor survival and that PCI may be an independent prognostic factor in neuroblastoma. In conclusion, immunostaining of PCNA is considered to be an appropriate index of proliferation activity in neuroblastoma. It relates well with tumor growth and other prognostic factors, and a high PCI is significantly related with a poor survival rate. PCI may be useful for providing prognostic information regarding neuroblastoma. Further studies are required, but our preliminary data suggest that PCNA immunoreactivity may be useful in identifying a poor prognostic group of patients. This may allow this group to receive more aggressive adjuvant therapy in the hope of producing more worthy survival rates.
REFERENCES 1. Hall PA, Levison DA: Review: Assessment of cell proliferation in histological material. J Clin Pathol 43:184-192, 1990 2. Hall PA, Wood AL: Immunohistochemical markers of cellular proliferation: Achievements, problems and prospects. Cell Tissue Kinet 23:531-549, 1990 3. Takasaki Y, Deng JS, Tan EM: A nuclear antigen associated with cell proliferation and blast transformation: Its distribution in synchronized cells. J Exp Med 154:1899-1909, 1981 4. Mathews MB, Bernstein RM, Franza BR, et al: Identity of the proliferating cell nuclear antigen and cyclin. Nature 309:374-376, 1984 5. Celis JE, Celis A: Cell cycle dependent variations in the distribution of the nuclear protein cyclin proliferating cell nuclear antigen in cultured cells: Subdivision of S phase. Proc Natl Acad Sci USA 82:3262-3266, 1985 6. Bravo R, Frank R, Blundell PA, et al: Cyclin/PCNA is the auxiliary protein of DNA polymerase-delta. Nature 326:515-517, 1987 7. Prelich G, Tan CK, Kostula M, et al: Functional identity of proliferating cell nuclear antigen and a DNA polymelase-0 auxiliary protein. Nature 326:517-520, 1987 8. Jaskulski D, DeRiel JK, Mercer WE, et al: Inhibition of cellular proliferation by antisense oligodeoxynucleotides to PCNA/ cyclin. Science 240:1544-1546, 1988 9. Waseen NH, Lane DP: Monoclonal antibody analysis of the proliferating cell nuclear antigen (PCNA). Structural conservation and the detection of a nuclear form. J Cell Sci 96:121-129, 1990 10. Garcia RL, Coltrera MD, Gown AM: Analysis of proliferative grade using anti-PCNA/cyclin monoclonal antibodies in fixed, embedded tissues. Am J Pathol 134:733-739, 1989 11. Hall PA, Levinson DA, Wood AL, et al: Proliferating cell nuclear antigen (PCNA) immunolocalisation in paraffin sections: An index of cell proliferation with evidence of deregulated expression in some neoplasms. J Pathol 162:285-294, 1990 12. Shimada H, Chattern J, Newton WA, et al: Histopatholigic prognostic factors in neuroblastic tumors; definition of subtypes of ganglioneuroblastoma and an age-linked classification of neuroblastomas. J Natl Cancer Inst 73:405-416, 1984 13. Hedley DW, Friedlander ML, Taylor IW, et al: Method for analysis of cellular DNA content of paraffin-embedded pathologi-
cal material using flow cytometry. J Histochem Cytochem 31:13331335, 1983 14. Gardes J, Lemke H, Baisch H, et al: Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133:1710-1715, 1983 15. Matsuno Y, Mukai K: Proliferating cell nuclear antigen (PCNA) Byori-To-Rinsyo 9:879-883, 1991 16. Smetana K, Gyokey F, Chan PK, et al: Proliferating cell nuclear antigen (PCNA) and human malignant tumor nuclear antigens (HMTNA) in nucleoli of human hematological malignancies. J Exp Clin Hemato146:133-141, 1983 17. Robbins BA, Vega D, Ogana K, et al: Immunohistochemical detection of proliferating cell nuclear antigen in solid human malignancies. Arch Pathol Lab Med 111:841-845, 1987 18. Benjamin DR, Gown AM: Aberrant cytoplasmic expression of proliferating cell nuclear antigen in Hodgkin's disease. Am J Surg Pathol 15:764-768, 1991 19. Woods AL, Hall PA, Shepherd NA, et al: The assessment of proliferating cell nuclear antigen (PCNA) immunostaining in primary gastrointestinal lymphomas and its relationship to histological grade, S + G2 + M phase fraction (flow cytometric analysis) and prognosis. Histopathology 19:21-27, 1991 20. Yu CCW, Hall PA, Fletcher CDM, et al: Haemangiopericytomas: The prognostic value of immunohistochemical staining with a monoclonal antibody to proliferating cell nuclear antigen (PCNA). Histopathology 19:29-33, 1991 21. Gansler T, Chatten J, Varello M, et al: Flow cytometric analysis of neuroblastoma: Correlation with histology and clinical outcome. Cancer 58:2453-2458, 1986 22. Oppedel BR, Storm-Mathisen I, Lie OS, et al: Prognostic factors in neuroblastoma: Clinical, histopathologic, and immunohistochemical features and DNA ploidy in relation to prognosis. Cancer 62:772-754, 1988 23. Taylor SR, Blatt J, Costantino JP, et al: Flow cytometric DNA analysis of neuroblastoma and ganglioneuroblastoma: A 10-year retrospective study. Cancer 62:749-754, 1988 24. Jain S, Filipe M, Hall PA, et al: Prognostic value of proliferating cell nuclear antigen in gastric carcinoma. J Clin Patho144:655-659, 1991
NFORMATION on cellular proliferation would be useful for predicting metastatic potential, recurrence, and overall prognosis in various human neoplasms.I, 2 Proliferating cell nuclear antigen (PCNA) is a cell-cycle-related nuclear protein that is maximally elevated in late G1 and S phase. 3-s It has recently been identified as the polymerase 0 accessory protein; therefore its level correlates directly with the rate of cellular proliferation and DNA synthesis. 6-8 A monoclonal antibody, PC10, has been shown to recognize PCNA in formalin-fixed, paraffin-embedded tissues. T M Neuroblastoma is the most common pediatric solid neoplasm. Although some biological variables such as N-myc gene amplification or DNA ploidy are reported to have relations with prognosis of this tumor, the relation between proliferating activity and prognosis is unknown. The purpose of this study is to analyze PCNA expression in human neuroblastoma using
I
From the Department of Pediatric Surgery, Osaka University Medical School, Osaka, Japan. Address reprint requests to Takaharu Oue, Department of Pediatric Surgery, Osaka University Medical School, 2-2 Yamadaoka, Suita City, Osaka 565, Japan. Copyright © 1995 by W.B. Saunders Company 0022-3468/95/3004-0004503. O0/ 0 528
immunohistochemical staining with PC10, and to assess whether it represents proliferating activity of neuroblastoma cells and to provide prognostic information. MATERIALS AND METHODS
Patient Profiles Samples of neuroblastoma were available from 67 patients who were operated on at Osaka University Hospital from 1970 to 1993. They were 36 boys and 31 girls, with a mean age of 2 years 2 months (from 14 days to 12 years of age). Eighteen of the patients were detected through mass screening. Histologically, 53 tumors were neuroblastoma and 14 were ganglioneuroblastoma. According to Evans' staging system, 13 patients were of stage I, 5 were of II, 18 were of III, 28 were of IV, and 3 were of IVs. Forty patients received chemotherapy before surgery.
Method Extirpated samples were immediately fixed with 10% formalin solution and processed with standard procedures for paraffin embedding. Conventional histological sections with 3 txm of thickness were cut and mounted on poly-L-lysine-coated glass slides. Immunostaining was performed with standard streptavidin-biotin immunoperoxidase method (LSAB kit; Dako Japan Corp; Kyoto, Japan). Sections were dewaxed in xylene, rehydrated through alcohol, and immersed in 3% hydrogen peroxide in methanol for 30 minutes to block endogenous peroxidase activity, and normal goat serum was applied for 10 minutes to reduce nonspecific antibody binding. The monoclonal antibody PC10, a murine IgG, was obtained from Dako corporation, which was used at a dilution of 20 times with 2 hours' incubation at room temperature. Biotinylated goat anti-mouse antibody was used as the linker molecule for 20 minutes, and sections were incubated in streptavidin-horseradish peroxidase complex for 10 minutes. Diamino-benzidine-hydrogen peroxidase was used to detect the presence of the peroxidase, and a light Mayer's hematoxylin counterstain was applied. To ensure consistency of PCNA staining, a known positive control was included in each round, and negative control was also included in which the primary antibody was replaced by phosphate-buffered saline (PBS).
Assessment of PCNA Staining To negate the tumor heterogeneity, at least five high-power (x 400) fields were randomly selected for cell counting. A minimum of 1000 tumor cells were counted in each case. Cells were considered positive for PCNA only when definite reddish-brown staining of the nucleus was identified. The PC10 labeling index, PCI, was defined as the percentage of positive tumor cell nuclei in the following way: Number of cells with positive staining PCI = Total number of tumor cells counted x 100 (%)
A s s e s s e d Variables Examinations were performed to determine whether there were significant relations between PCNA expression and various patho-
Journalof PediatricSurgery,Vol 30, No 4 (April), 1995:pp 528-532
PCNA IMMUNOSTAINING IN HUMAN NEUROBLASTOMA
logical and clinical variables such as mitotic karyorrhexis index, cell concentration, tumor weight and histological classification, age, clinical stage, local invasion, metastasis, and DNA ploidy. Mitotic karyorrhexis index (MKI) defined in Shimada's classification, 12 was determined by counting the number of mitotic figures and karyorrhexis in 5000 tumor cells. To analyze DNA ploidy, flow cytometry was performed using paraffin blocks as described by Hedley et al. 13 The groups were compared using unpaired t test. For multifactorial analysis, analysis of variance (ANOVA) test was applied. Survival curve according to PCNA expression was calculated using the Kaplan-Meyer method, and the Generalized Wilcoxon test was applied to compare the survival rates.
RESULTS
PCNA Staining Figure 1 shows the typical nuclear staining of samples. Positive staining of PCNA was confined to the nucleus of proliferating cells. PCI varied within a range of 0% to 75%.
Effect of Chemotherapy During these 10 years, we have performed aggressive chemotherapy before surgery using cisplatin and doxorubicin for stage III and IV neuroblastoma patients. To access the effect of chemotherapy in this group, PCNA expression was compared between the patients who received chemotherapy and those who did not (Fig 2). PCI was significantly low in patients who received aggressive chemotherapy before surgery. To exclude the effect of chemotherapy, patients who received chemotherapy before surgery were excluded and 40 patients were analyzed for the following study.
Relation Between Pathological Variables and PCI PCI shows significant positive correlation with mitotic karyorrhexis index (Fig 3), cell concentration
529
PCI 80 %70
0
P<0.05 0
6050 40
ol
30 20 10
I
' -I-
chemotherapy Fig 2. Relation between chemotherapy and PCI. To access the effect of chemotherapy in this group, PCNA expression was compared between the patients who received chemotherapy and those who did not in the stage III and IV patients during these 10 years. PCI was significantly low in the patients who received aggressive chemotherapy before surgery. (O), dead; (©), alive.
(Fig 4), and tumor weight (Fig 5). Histologically, PCI in ganglioneuroblastoma is significantly lower than that in neuroblastoma (Table 1). PCI is low when differentiated cells appeared (Fig 6).
Relation Between Clinical Variables and PCI
Fig 1. PCNA-positive staining in stage IV neuroblastoma. Positive staining of PCNA was confined to the nucleus of proliferating cells.
PCI was significantly high in the cases of stage III and IV in comparison with the cases of stage I, II and IV. Tumors invading to other organs or with metastatic lesions in lymph nodes or liver show an increasing percentage of PCNA-positive cells. PCI was low in the cases under 1 year of age but not significant. PCI was significantly low in aneuploid tumors compared with diploid (Table 1).
530
OUE ET AL
2000
MKI
g
400
y=4.882x+4.484, r= 0.925 /
O
O ~
y =13.86 x - 2.58 r = 0.591
1500
300 1000 200
500
100 PCI
0 0
20
40
60
0
~ 0
% 80
PCl
20
40
60 %
Fig 3. Relationship between PCI and MKI. There is significant correlation between PCI and MKI.
Fig 5, Relationship between PCI and tumor weight, There is significant correlation between PCI and tumor weight.
Prognosis
recurrence, and overall prognosis. 12 PCNA is a 36 kDa nuclear protein, and its level is maximally elevated in late G1 and S phase and expression is closely linked to the cell cycle.3-5 It has recently been identified as the polymerase 0 accessory protein, and its level therefore correlates directly with the rate of cellular proliferation and DNA synthesis. 6-8 A monoclonal antibody, PC10, has been shown to recognize PCNA in formalin-fixed, paraffin-embedded tissues. 911 Current methods of assessing the proliferation and growth of tumors from slide-based materials with monoclonal antibodies such as Ki-67 were limited because they require cryostat sections of freshfrozen material. 1,2J4 The major advantage of staining with PC10 is its detectability in formalin-fixed paraffinembedded tissues, which enables us to perform retro-
Figure 7 shows survival cUrves that show the relationship between prognosis and PCI. Patients with PCI higher than 30% have a worse survival rate compared with that of patients whose PCI is lower than 10% (P < .01). These results suggest that a high PCI significantly related with poor survival and that PCI may be an independent prognostic factor in neuroblastoma. DISCUSSION Information on cell kinetics would be useful in understanding tumor behavior. In a variety of malignant neoplasms, correlations have been noted between proliferative activity and metastatic potential,
cell density 2.0 . . . . . . . .
rate of differentiated cells 30 . . . . . % o 25 o
' ....
/100~tm2 1.5
OOo
O
20
o e
1.0
~1~
15
°.4o°
10
0.5 0
0
O
+ °y=O.O2x r = 0.667
l 0
20
40
5
0.548
o
PCI 60 %
80
Fig 4. Relationship between PCI and cell concentration. There is significant correlation between PCI and cell density,
0
o
o o~ig ® ~ o
0
20
o
40
o
.
PCl
6O %
Fig 6. Relationship between cell differentiation and PCI. PC! is low in the tumor in which the rate of differentiated cells is higher than 5%.
PCNA IMMUNOSTAINiNG IN HUMAN NEUROBLASTOMA
531
Table 1. Relation Between Prognostic Variables and PCI
PCI Variables
Age < 1 yr >1 yr Stage I, II, IV III, IV Histology NB GNB Local invasion C1 C2 C3 Lymph node metastasis
n
(mean +- SD)
Probability
26 14
11.6 -+ 10.0 21.6 -+ 6.7
.078
19 21
7.4 -+ 9.9 22.1 -+ 19.5
.0051
32 7
18.4 - 17.6 0.35 -+ 0.41
.011
14
5.6 -+ 7,4 8.8 _+ 13.4 25,5 -+ 19.1
.05 (Cl versus C3)
7 18
n0, nl
24 13
8.5 -+ 10.4 27.0 _+ 22.1
.0014
n2, n 3
26 6
15.2 _+ 15.1 33.7 _+ 20.8
.017
10 17
30.5 -+ 25.0 14.1 -+ 9.7
.023
Liver metastasis + DNA ploidy diploid aneuploid
Abbreviations: NB, neuroblastoma; GNB, ganglioneuroblastoma; C~, localized to the capsule; C2, invasion to near organs; C3, invasion to opposite site; no, no metastasis; nl, local metastasis; n2, n3, distant metastasis.
spective and morphological evaluation on cellular proliferation. On the contrary, there are some problems in assessing proliferating activity using paraffin blocks. A partial loss of antigenicity of PCNA after formalin fixation cannot be excluded. Matsuno and Mukai 15showed that PCNA immunoreactivity is markedly reduced over 48 hours' formalin fixation, so materials should not be fixed longer than 48 hours. In our institute, all pathological materials were embedded and stained in our own pathological laboratory within 48 hours. *PCI<10 n=18
100 %
0,PO,
80
0
60 40
*30<_--PCI n=5 20 0
* P<0.01
1
2
4
year
Fig 7. Survival curve according to PCNA expression. Patients with PCI higher than 30% have a worse survival rate compared with these o PCI (P < ,01), with a lower than 10Vo
Recent reports suggest that this antibody may be useful in leukemias, lymphomas, and certain adult tumors as an independent prognostic variable,26-2° but no pediatric solid tumors have been studied in detail. In this study, we analyzed PCNA expression in human neuroblastoma using immunohistochemical techniques and assessed whether it represented proliferating activity and provided prognostic information. In order to assess the proliferating activity, Shimada et al defined the MKI, which is the number of mitotic figures and karyorrhexis in 5000 tumor cells, and applied it in their histological classification of neuroblastomas. 12 According to this classification, tumors with high MKI were classified as "unfavorable histology." In our study, PCI showed significant positive correlation with MKI. This result may indicate that PC! is considered to be an appropriate index of proliferation activity in neuroblastoma. PCI also showed significant positive correlation with tumor weight and clinical stage. Tumors with high proliferating activity may grow rapidly, so that tumor size may become larger and clinical stage may be advanced at the time of diagnosis. PCI was high in the tumors with local invasion and in the tumors with metastatic lesions in lymph nodes or liver. These results may indicate that proliferating activity plays an important role in both invasion and metastasis in neuroblastomas. Histologically, PCI in ganglioneuroblastoma was significantly lower than that in neuroblastoma. PCI was low in the tumors in which the rate of differentiated cells was higher than 5% (Fig 6). From these findings, differentiated cells are considered to have very low proliferating activity. Recent studies of DNA ploidy in neuroblastoma show that prognosis of the patients with aneuploid tumor is significantly good compared with the patients with diploid tumor. 21-23 In our study, PCI was significantly low in aneuploid tumors, which are considered to have poor proliferating activity or good prognosis in neuroblastoma. Woods et a119 showed that there is a significant relationship between PCI and S + G2M phase fraction as measured by flow cytometric analysis in gastrointestinal lymphomas. To the contrary, Jain et a124 showed that no significant correlation was observed in gastric carcinoma. We are investigating the relationship between PCI and S + G2M phase fraction in neuroblastoma, but sufficient correlation is not available (data not shown). One of the reasons is that it is difficult to recognize the S + G2M phase fraction in aneuploid tumors.
532
OUE ET AL
During these 10 years, we have performed aggressive chemotherapy using cisplatin and doxorubicin for advanced neuroblastoma patients. To access the effect of chemotherapy on the proliferative activity, PCNA expression was analyzed in the stage III and IV patients who had undergone surgery since 1980. The results show that proliferating activity was reduced by chemotherapy. Among chemotherapy groups, the patients whose tumor still showed a high percentage of PCI had a poor prognosis (Fig 1). These findings show that PCI may be used to access the effectiveness of chemotherapy and to predict the prognosis of a patient who received preoperative chemotherapy. Patients with a PCI higher than 30% have a worse survival rate compared with those who have a PCI
lower than 10%. These results suggest that a high PCI is significantly related to poor survival and that PCI may be an independent prognostic factor in neuroblastoma. In conclusion, immunostaining of PCNA is considered to be an appropriate index of proliferation activity in neuroblastoma. It relates well with tumor growth and other prognostic factors, and a high PCI is significantly related with a poor survival rate. PCI may be useful for providing prognostic information regarding neuroblastoma. Further studies are required, but our preliminary data suggest that PCNA immunoreactivity may be useful in identifying a poor prognostic group of patients. This may allow this group to receive more aggressive adjuvant therapy in the hope of producing more worthy survival rates.
REFERENCES 1. Hall PA, Levison DA: Review: Assessment of cell proliferation in histological material. J Clin Pathol 43:184-192, 1990 2. Hall PA, Wood AL: Immunohistochemical markers of cellular proliferation: Achievements, problems and prospects. Cell Tissue Kinet 23:531-549, 1990 3. Takasaki Y, Deng JS, Tan EM: A nuclear antigen associated with cell proliferation and blast transformation: Its distribution in synchronized cells. J Exp Med 154:1899-1909, 1981 4. Mathews MB, Bernstein RM, Franza BR, et al: Identity of the proliferating cell nuclear antigen and cyclin. Nature 309:374-376, 1984 5. Celis JE, Celis A: Cell cycle dependent variations in the distribution of the nuclear protein cyclin proliferating cell nuclear antigen in cultured cells: Subdivision of S phase. Proc Natl Acad Sci USA 82:3262-3266, 1985 6. Bravo R, Frank R, Blundell PA, et al: Cyclin/PCNA is the auxiliary protein of DNA polymerase-delta. Nature 326:515-517, 1987 7. Prelich G, Tan CK, Kostula M, et al: Functional identity of proliferating cell nuclear antigen and a DNA polymelase-0 auxiliary protein. Nature 326:517-520, 1987 8. Jaskulski D, DeRiel JK, Mercer WE, et al: Inhibition of cellular proliferation by antisense oligodeoxynucleotides to PCNA/ cyclin. Science 240:1544-1546, 1988 9. Waseen NH, Lane DP: Monoclonal antibody analysis of the proliferating cell nuclear antigen (PCNA). Structural conservation and the detection of a nuclear form. J Cell Sci 96:121-129, 1990 10. Garcia RL, Coltrera MD, Gown AM: Analysis of proliferative grade using anti-PCNA/cyclin monoclonal antibodies in fixed, embedded tissues. Am J Pathol 134:733-739, 1989 11. Hall PA, Levinson DA, Wood AL, et al: Proliferating cell nuclear antigen (PCNA) immunolocalisation in paraffin sections: An index of cell proliferation with evidence of deregulated expression in some neoplasms. J Pathol 162:285-294, 1990 12. Shimada H, Chattern J, Newton WA, et al: Histopatholigic prognostic factors in neuroblastic tumors; definition of subtypes of ganglioneuroblastoma and an age-linked classification of neuroblastomas. J Natl Cancer Inst 73:405-416, 1984 13. Hedley DW, Friedlander ML, Taylor IW, et al: Method for analysis of cellular DNA content of paraffin-embedded pathologi-
cal material using flow cytometry. J Histochem Cytochem 31:13331335, 1983 14. Gardes J, Lemke H, Baisch H, et al: Cell cycle analysis of a cell proliferation-associated human nuclear antigen defined by the monoclonal antibody Ki-67. J Immunol 133:1710-1715, 1983 15. Matsuno Y, Mukai K: Proliferating cell nuclear antigen (PCNA) Byori-To-Rinsyo 9:879-883, 1991 16. Smetana K, Gyokey F, Chan PK, et al: Proliferating cell nuclear antigen (PCNA) and human malignant tumor nuclear antigens (HMTNA) in nucleoli of human hematological malignancies. J Exp Clin Hemato146:133-141, 1983 17. Robbins BA, Vega D, Ogana K, et al: Immunohistochemical detection of proliferating cell nuclear antigen in solid human malignancies. Arch Pathol Lab Med 111:841-845, 1987 18. Benjamin DR, Gown AM: Aberrant cytoplasmic expression of proliferating cell nuclear antigen in Hodgkin's disease. Am J Surg Pathol 15:764-768, 1991 19. Woods AL, Hall PA, Shepherd NA, et al: The assessment of proliferating cell nuclear antigen (PCNA) immunostaining in primary gastrointestinal lymphomas and its relationship to histological grade, S + G2 + M phase fraction (flow cytometric analysis) and prognosis. Histopathology 19:21-27, 1991 20. Yu CCW, Hall PA, Fletcher CDM, et al: Haemangiopericytomas: The prognostic value of immunohistochemical staining with a monoclonal antibody to proliferating cell nuclear antigen (PCNA). Histopathology 19:29-33, 1991 21. Gansler T, Chatten J, Varello M, et al: Flow cytometric analysis of neuroblastoma: Correlation with histology and clinical outcome. Cancer 58:2453-2458, 1986 22. Oppedel BR, Storm-Mathisen I, Lie OS, et al: Prognostic factors in neuroblastoma: Clinical, histopathologic, and immunohistochemical features and DNA ploidy in relation to prognosis. Cancer 62:772-754, 1988 23. Taylor SR, Blatt J, Costantino JP, et al: Flow cytometric DNA analysis of neuroblastoma and ganglioneuroblastoma: A 10-year retrospective study. Cancer 62:749-754, 1988 24. Jain S, Filipe M, Hall PA, et al: Prognostic value of proliferating cell nuclear antigen in gastric carcinoma. J Clin Patho144:655-659, 1991