Correlation p53 expression and human papilloma virus deoxyribonucleic acid with clinical outcome in early uterine cervical carcinoma

Cancer Detection and Prevention 29 (2005) 528–536 www.elsevier.com/locate/cdp

Correlation p53 expression and human papilloma virus deoxyribonucleic acid with clinical outcome in early uterine cervical carcinoma Akiko Ikuta MD*, Junko Saito MD, PhD, Tomomi Mizokami MD, Tsuyoshi Nakamoto MD, Masahiro Yasuhara MD, Fumie Nagata MD, Mayumi Nakajima MD, Izumi Matsuo MD, Katsuhiko Yasuda MD, PhD, Hideharu Kanzaki MD, PhD Department of Obstetrics and Gynecology, Kansai Medical University, 10–15 Fumizono-cho, Moriguchi Osaka 570-8507, Japan Accepted 15 September 2005

Abstract Background: In the present study we assessed whether expression of p53 protein or HPV DNA correlates with recurrence as well as several known prognostic factors in uterine cervical carcinoma. Methods: Forty-nine patients with FIGO stage IA–IIB who underwent hysterectomy between 1998 and 2002 were retrospectively studied. All 49 cancer tissue samples were used for immunohistochemical study. Twenty-five of 49 cases were also examined by PCR–RFLP for detection and typing of HPV DNA. Results: Twenty of 49 (40.8%) specimens demonstrated nuclear staining for p53. A significant association between p53 overexpression and age, hormonal status, FIGO stage, or recurrence was observed ( p = 0.02, 0.01, 0.03, 0.01). However, no significant association was found between p53 overexpression and lymph node metastases, parametrium involvement, or risk of death ( p = 0.18, 0.06, 0.14). Nineteen of 25 (76%) were HPV DNA-positive and 6 (24%) were negative. Discussion: There was no relation between HPV DNA positivity and age, FIGO stage, lymph node metastases, parametrium involvement, recurrence, or risk of death. Conclusion: p53 overexpression is associated with age, hormonal status, FIGO stage, and recurrence in uterine cervical carcinoma. # 2005 International Society for Preventive Oncology. Published by Elsevier Ltd. All rights reserved. Keywords: Uterine cervical carcinoma; p53; Immunohistochemical study; Human papilloma virus (HPV); Prognosis; Recurrence; Metastases; Clinical outcome; Hormonal status; HPV DNA

1. Introduction Uterine cervical carcinoma is one of the most prevalent forms of carcinoma found in women worldwide, accounting for almost 12% of all carcinomas in women [1]. The development and progression of carcinoma are considered multi-step processes in which host genetic as well as environmental factors may be involved [2–4]. Several epidemiologic studies have implicated human papilloma virus (HPV) infection, especially HPV types 16 and 18, in * Corresponding author. Tel.: +81 6 6992 1001; fax: +81 6 6992 3438. E-mail address: [email protected] (A. Ikuta).

carcinogenesis of uterine cervical carcinoma. Furthermore, HPV types 31, 33, 35, 39, 45, 52, and 58 have been detected in this carcinoma, though less frequently [5–7]. The most commonly identified carcinoma-related gene alteration, however, is mutation of the p53 tumor suppressor gene [2,4,8–10]. The p53 gene encodes a 53 kDa nuclear phosphoprotein (p53) involved in cell cycle regulation, and normal cells containing wild-type p53 results in cell cycle arrest, leading either to DNA repair or apoptotic cell death [2,4,8–10]. It has also been shown that genes that have undergone mutations can function not only as tumor suppressors but also as dominant transforming genes eliciting malignant transformation, similar to proto-oncogene

0361-090X/$30.00 # 2005 International Society for Preventive Oncology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.cdp.2005.09.002

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536

[2,4,8–10]. The loss of wild-type p53 protein or mutation of this protein may lead to oncogenic activity [2,4,8–13]. In addition to carcinogenesis, several studies have also reported an association between p53 mutation and tumor progression, metastasis, or prognoses in human solid malignancies, including gynecologic carcinoma [2,10,11, 14,15]. In particular, the E6 protein of the high-risk HPV forms a complex with p53, resulting in degradation of p53 protein [3,9,10,16–20]. This in turn obviates the arrest of the cell cycle at the G1/S interphase, allowing the continued replication of the DNA damaged cells, leading to genome instability and accumulation of mutations [3,9,10,16–20]. p53 protein is unstable and has a half-life of only 20– 30 min, while the mutant p53 protein is more stable and has a prolonged half-life [8,10,21]. Consequently, expression of p53 protein is easily detectable by immunohistochemical techniques [8,10,21]. In uterine cervical carcinoma, prognostic factors such as clinical stage, lymph node metastases, or parametrium involvement have been established [22,23]. We need new prognostic factors, however, to identify patients at low-risk of recurrence who could be offered a more conservative therapy and patients at high-risk of recurrences who should be offered more intensive treatment. In this study, we investigated the relation between p53 overexpression and HPV DNA and disease outcome.

2. Materials and methods 2.1. Patients The study was performed on primary uterine cervical carcinoma tissue samples from 49 patients undergoing hysterectomy at the Department of Obstetrics and Gynecology, Kansai Medical University, between 1998 and 2002. Patients had received neither radiotherapy nor chemotherapy before surgery. Postoperatively, adjuvant radiotherapy was administrated in cases with lymph node metastases, parametrial involvement, or marked vascular space infiltration. Forty-nine patients completed the International Federation of Gynecology and Obstetrics (FIGO) standard treatment: 20 (40.8%) to surgery alone, 8 (16.3%) to surgery plus radiotherapy, 18 (36.7%) to chemoradiation therapy, and 3 (6.1%) to chemotherapy. The median age of patients was 45 years, ranging from 30 to 90 years. Twenty-one patients were aged 39 years or younger, and 28 were aged 40 years or older. Thirty-one patients were premenopausal status, and 18 were postmenopause. The clinical stage of uterine cervical carcinoma was determined by FIGO classification and histologically assessed according to the World Health Organization (WHO) criteria. Patients with cervical carcinoma included 8 (16.3%) in stage IA, 29 (59.2%) in stage IB, 1 (2%) in stage IIA, and 11 (22.4%) in stage IIB, and the histology included 36 patients (73.5%) with squamous cell carci-

529

noma, 11 (22.4%) with adenocarcinoma, and 2 (4.1%) with adenosquamous carcinoma. All tissue samples were used for immunohistochemical examination, while 25 cancerous tissue samples were examined for HPV identification and typing. Specimens obtained at surgery were immediately frozen in liquid nitrogen and stored at
80 8C until DNA extraction or Western blot analysis. DNA from specimens was extracted according to the standard SDS–proteinase K procedure [19]. 2.2. Immunohistochemistry for p53 protein p53 was detected with the mouse monoclonal antibody p53, clone DO-7, which recognizes wild-type and mutant forms of the p53 protein. All tissue blocks had been fixed in 10% formalin and embedded in paraffin. The specimens were classified based on hematoxylin and eosin-stained sections using conventional histologic criteria. Furthermore, 3 mm thick sections were cut serially and mounted on 3aminopropyltriethoxy-silane-coated glass slides for immunohistochemical studies. The streptavidin–biotin complex peroxidase (SAB) method (Histofine SAB-PO kit; NICHIREI Corp., Tokyo, Japan) was used. Briefly, the tissue sections were first deparaffinized in xylene, then rehydrated in a graded alcohol series. Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide in methanol. The sections were then washed in 0.01 M phosphate-buffered saline (PBS, pH 7.2), and a microwave antigen retrieval procedure using citrate buffer (pH 6.0) was performed. After microwave exposure, the slides were allowed to cool to room temperature then incubated for 15 min in 10% normal horse serum prior to the application of the primary antibody. Incubation with a mouse monoclonal antibody against p53 (DO-7 pre-diluted antibody; NICHIREI Corp., Tokyo, Japan) was performed for 60 min at room temperature. It is known that the E6 protein of the high-risk HPV forms a complex with p53, resulting in degradation of p53 protein [3,9,10,16–20]. Therefore, as a control a proliferation marker, immunohistochemical staining for Ki-67 (pre-diluted mouse monoclonal antibody; Zymed Laboratories Inc., USA) was performed for 3 h at room temperature. Antibody staining was developed with diaminobenzidine (DAB) for 10 min. The sections were then counterstained with hematoxylin, dehydrated in a graded alcohol series, cleared in xylene, and finally mounted in balsam. Negative control sections were treated identically, except that the specific antibodies were replaced by normal mouse serum. The sections of ovarian carcinomas known to express the investigated antigen were used as a positive control. p53 overexpression was evaluated under a double-headed microscope at 200-fold magnification. Five fields were chosen at random. Tumor cells on each section were counted to determine the percentage of positive cells. Staining was considered positive when more than 10% of the cells were nuclear-stained. These evaluations were made without

530

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536

knowledge of the patient’s clinicopathological characteristics or clinical outcome. 2.3. Typing of HPV DNA To ensure adequate DNA preparation, PCR amplification of beta globin was previously performed in a separate reaction using primers PCO3 (50 -ACA CAA CTG TGT TCA CTA GC-30 ) and PCO4 (50 -CAA CTT CAT CCA CGT TCA CC-30 ) with a product size of 110 bp as a control [24,25]. PCR was performed in a Takara PCR thermal cycler (Takara shuzo Co. Ltd., Japan) for 40 cycles of denaturation (94 8C, 1.0 min), annealing (40 8C, 2.0 min), and extension (72 8C, 1.5 min). Each reaction mixture (50 ml) contained the following: KCl (50 mM), Tris–HCl pH 8.3 (10 mM), MgCl2 (3.5 mM), dNTPs (200 mM each), primers PCO3 and PCO4 (50 pmol each), Taq polymerase (1 units), and pre-treated DNA solution (10 ml). In the present study, the reaction products were amplified by PCR using consensus primers for the L1 region (L1C1; 50 -CGT AAA CGT TTT CCC TAT TTT TTT-30 , L1C2; 50 -TAC CCT AAA TAC TAT GTATTG30 , and L1C2M; 50 -TAC CCT AAA-TAC CCT ATA TTG-30 ) for 40 cycles of denaturation (95 8C, 1.5 min), annealing (48 8C, 1.5 min), and extension (70 8C, 2.0 min), on the same thermal cycler. Because L1C2M increases the sensitivity of L1-PCR to HPV type 58 without decreasing sensitivity to other HPV types, this primer was added [24]. Each reaction mixture (100 ml) contained the following: KCl (50 mM), Tris–HCl pH 8.3 (10 mM), MgCl2 (1.5 mM), dNTPs (250 mM each), primer 1 (1 mM), primer 2, 3 (0.5 mM each), Taq polymerase (5 U), and DNA solution (1 mg). The negative control consisted of the reaction mix without DNA. After the reaction, 10 ml of the product was electrophoresed through 4% Nu SieveTM GTG agarose (Takara Shuzo Co. Ltd., Japan) gel, final stained with 1 mg/ ml ethidium bromide; the gel was then photographed under UV light. The length between these consensus primers is 244 bp in HPV types 6 and 11, 253 bp in types 16 and 18, and 256 bp in types 31 and 33 [24]. Specimens that were negative and untyped HPV (HPV X) with consensus primers were analyzed with type-specific primers (HPV detection set and HPV typing set; Takara Co. Ltd., Japan). The purpose of this primers was to maximize the identification of other HPV types not detected by the L1 consensus primers. In this case, PCR was 30 cycles of denaturation (94 8C, 0.5 min), annealing (55 8C, 2.0 min), and extension (72 8C, 2.0 min). The length between the former primers is 140 bp in HPV types 16 and 18, and 141 bp in HPV type 33 for the E6 region. DNA detection was based upon the amplification by PCR, and was confirmed by restriction fragment length polymorphism analysis. Three kinds of restriction enzymes, Rsa I, Dde I, and Hae III (Takara Shuzo Co. Ltd., Japan), were used as primary screening enzymes in this study. The HPV types were identified by restriction fragment length polymorphism (RFLP) [26].

2.4. Western blotting Protein from tissue samples was extracted using T-PERTM tissue protein extraction reagent (PIERCE, Rockford, USA), and then homogenized. Two micrograms of proteins was electrophoresed through 10% sodium dodecyl sulfate (SDS) polyacrylamide gel and transferred to a cellulose membrane. The membranes were blocked with 5% skim milk and immunoreacted with primary mouse anti-human p53 monoclonal antibody (DO-7) (Novocastra, Newcaslte, UK) at 1:1000 dilution for 60 min at room temperature. Breast and colonic carcinomas known to express the immunostaining for p53 protein were used as a positive control. Negative controls were used ovarian carcinomas without p53 overexpression. After washing, the membrane was incubated with anti-mouse IgG. Bound antibodies were detected using the LumiGLO chemiluminescent system (KPL Inc., MD, USA). In this study, western blotting could be performed in only one case, because other samples were very small. 2.5. Statistical analysis Comparison of groups was performed using Fisher’s exact probability test or x2-test. A value of p < 0.05 was considered significant. Recurrence-free survival or overall survival according to p53 overexpression was constructed using Kaplan–Meier method, and the curves were compared using the log-rank test.

3. Results 3.1. Immunohistochemical staining of p53 and clinicopathological parameters Positive staining for p53 overexpression was localized in the nuclei of carcinoma cells (Fig. 1). Twenty of 49 (40.8%) cervical carcinomas demonstrated nuclear staining for p53.

Fig. 1. Immunohistochemical staining for p53 in squamous cell carcinoma. Several nuclei of tumor cells show overexpression (at 200 magnification).

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536

Fig. 2. Western blot analysis of cervical carcinoma using DO-7 antibody. Overexpression of p53 protein is observed. A major band at 53 kd can be seen. Lane 1, cervical carcinoma tissue sample; lane 2, positive control.

531

In addition, p53 protein expression was observed by Western blot analysis as shown in Fig. 2. While, positive staining of Ki-67 was observed in 40 (81.6%). Association between p53 overexpression and clinicopathological parameters is shown in Table 1. Twenty of 49 (40.8%) cervical carcinomas demonstrated nuclear staining for p53, including 32.4% in stage IA/IB and 66.7% in stage IIA/IIB ( p = 0.03). p53 overexpression was shown in 23.8% of the younger group (<40 years) and 53.6% of the older group (40 years ) ( p = 0.02). Thirty-one patients were premenopausal status, and 18 were postmenopause. p53 overexpression was detected in 29.0% of the premenopausal group and 66.7% of the postmenopausal group ( p = 0.01). The distribution of p53 overexpression according to the risk of recurrence was 31.6% in patients with no recurrence and 72.7% in those with recurrence ( p = 0.01). The recurrence-free survival curves in relation to p53 overexpression are shown in Fig. 3(A). p53 overexpression was found to be a significant for recurrence-free survival by the log-rank survival analysis (x2 = 7.62, p = 0.005). Though there was no significant, the expression of p53 was detected in three of four death patients. There was a significant association between p53 overexpression and age, hormonal status or FIGO stage ( p = 0.02, 0.01, 0.03), as well as a significant association

Table 1 Association between p53 overexpression and clinicopathological parameters Clinicopathological parameters

No. of patients

No.

p53 over expression %

p-Value

Age (years) <40 40

23 28

5 15

23.8 53.6

0.02

Hormonal status Premenopause Postmenopause

31 18

9 12

29.0 66.7

0.01

FIGO stage IA/IB IIA/IIB

37 (8/29) 12 (1/11)

12 (2/10) 8 (1/7)

32.4 66.7

0.03

Histology SCC AC ASC

36 11 2

17 3 0

47.2 27.3 0

0.24

Lymph node metastases No 22 Yes 17 unknown 10

7 9 4

30.4 56.3 36.4

0.18

Parametrium involvement No 32 Yes 17

10 10

31.2 58.8

0.06

Recurrence No Yes

38 11

12 8

31.6 72.7

0.01

Death No Yes

45 4

17 3

37.7 75

0.14

SCC: squamous cell carcinoma; AC: adenocarcinoma; ASC: adenosquamous carcinoma.

Fig. 3. Kaplan–Meier curves for recurrence-free survival (A) or overall survival (B) according to p53 overexpression in patients with cervical carcinoma (log-rank test, A; x2 = 7.62, d.f. = 1 and p = 0.005, B; x2 = 2.56, d.f. = 1 and p = 0.1). When p53 overexpression was observed, survival rate was not significant (B), but recurrence rate increased significantly (A).

532

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536

between p53 overexpression and recurrence in the present study ( p = 0.01). The overexpression of p53 in squamous cell carcinoma (47.2%) was higher than that (13%) in either adenocarcinoma or adenosquamous carcinoma. However, the difference was not significant ( p = 0.24). There was no significant association between p53 overexpression and lymph node metastases ( p = 0.18). Although p53 overexpression was associated with a greater frequency of parametrium involvement, this association was not significant ( p = 0.06). The risk of death from cervical carcinoma was increased in patients with p53 overexpression, although the difference was not significant ( p = 0.14). The overall survival curves in relation to p53 overexpression are shown in Fig. 3(B) . Although p53 overexpression is poorer than that of the p53 negative expression, no significant difference was found (x2 = 2.56, p = 0.1). Because the length of follow-up ranged from 21 to 71 months, 5-year survival is not yet available. 3.2. Detection of HPV DNA and clinicopathological parameters Twenty-five of 49 cases were used for detection of HPV DNA by PCR–RFLP. Nineteen of 25 (76%) were HPV Table 2 Association between HPV DNA and clinicopathological parameters Clinicopathological parameters

No. of patients

HPV DNApositive case No.

p-Value %

Age (years) <40 40

6 19

4 15

66.7 78.9

0.54

Hormonal status Premenopause Postmenopause

12 13

6 11

50 84.6

0.07

FIGO stage IA/IB IIA/IIB

15 (1/14) 10 (1/9)

12 (0/12) 7 (0/7)

80 70

0.56

Histology SCC AC ASC

20 5 0

15 4 0

75 80 0

0.81

Lymph node metastases No 9 Yes 13 Unknown 3

6 11 2

66.7 84.6 66.7

0.32

Fig. 4. The electropheretic pattern of untyped HPV (HPV X) by L1 consensus primers. HPV DNA fragments were amplified, but the restriction map did not confirm: 1, uncut; 2, Rsa I; 3, Dde I; 4, HaeIII; positive control, SiHa cell including HPV 16. The molecular markers (øx 174 DNA digested with Hae III) are shown on the left; bp (base pair).

DNA-positive and 6 (24%) were negative in the present study. Sixteen of 19 were detected by PCR–RFLP, and three could be amplified by HPV detection set. HPV DNA was not amplified at all by HPV typing set. The association between HPV DNA-positive cases and clinicopathological parameters is summarized in Table 2. There was no association between HPV-positive cases and age, FIGO stage, histology, lymph node metastases, parametrium involvement, recurrence, and risk of death. Although HPV-positive cases were slight associated with postmenopausal status, this association was not significant ( p = 0.07). 3.3. Relation between p53 overexpression and HPV typing Of the 19 HPV-positive cases, 9 (47.4%) demonstrated HPV type 16, 1 (5.3%) HPV type 18, 2 (10.5%) HPV type 52, and 7 (36.8%) demonstrated unknown types (HPV X) (Fig. 4). There was no association between p53 overexpression and detection or type of HPV DNA, as shown in Table 3. Table 3 Relationship between p53 overexpression and HPV DNA or type of HPV

Parametrium involvement No 11 Yes 14

8 11

72.7 78.6

0.73

Recurrence No Yes

17 8

13 6

76.5 75

0.93

Death No Yes

22 3

16 3

72.7 100

0.30

SCC: squamous cell carcinoma; AC: adenocarcinoma; ASC: adenosquamous carcinoma.

HPV case HPV HPV HPV HPV

p53 overexpression (
) (n = 11) No. (%)

p53 overexpression (+) (n = 14) No. (%)

p-Value

DNA-positive

8 (72.7)

11 (78.6)

0.73

16 18 52 X

4 (36.4) 1 (9.1) 0 3 (27.3)

5 (35.7) 0 2 (14.3) 4 (28.6)

0.84 0.22 0.2 0.95

X: unclassified HPV types.

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536 Table 4 Relationship between recurrence and clinicopathological parameters in patients with cervical carcinoma Clinicopathological parameters

No. of patients

Patients with recurrence

p-Value

No.

%

21 28

3 8

14.2 28.6

0.23

Hormonal status Premenopause Postmonopause

31 18

6 5

19.3 27.8

0.84

FIGO stage IA/IB IIA/IIB

37 (8/29) 12 (1/11)

7 (1/6) 4 (1/3)

18.9 33.3

0.29

Histology SCC AC ASC

36 11 2

8 3 0

22.2 27.3 0

0.24

Lymph node metastases No 22 Yes 17 Unknown 10

2 5 4

9.1 29.4 40

0.07

Parametrium involvement No 32 Yes 17

2 9

6.3 52.9

0.0001

Age (years) <40 40

SCC: squamous cell carcinoma; AC: adenocarcinoma; ASC: adenosquamous carcinoma.

3.4. Relation between recurrence and clinicopathological parameters There was a highly significant association between recurrence and parametrium involvement, as shown in Table 4 ( p = 0.0001). The rate of recurrence was higher in the older group (40 years ) than the younger group (<40 years), though this difference was not significant ( p = 0.23). Recurrence occurred with slightly greater frequency in patients with a more advanced clinical stage. There was no association between recurrence and hormonal status ( p = 0.84) or histology ( p = 0.24). Lymph node metastases

533

were associated with a greater frequency of recurrence. However, this difference was not significant ( p = 0.07). In the present study, recurrence was observed in 11 patients (22.4%) and death in four (8.1%). Table 5 summarizes clinicopathological parameters, p53 overexpression, HPV DNA or typing, or disease-free survival in patients with recurrence. The median age at recurrence was 55 years, ranging from 33 to 90 years. Disease-free survival ranged from 5 to 32 months, and the median period was 10 months. In particular, p53 overexpression was observed in 72.7% (8/11).

4. Discussion In uterine cervical carcinoma, clinical stage, lymph node metastases, or parametrium involvement are important prognostic factors [22,23]. p53 plays an important role in regulating DNA repair or apoptosis [2,4,8–10]. With regard to the prognostic significance of p53 overexpression in cervical carcinoma, some published data are controversial [27–35]. Our immunohistochemical study showed that p53 overexpression is associated with age, hormonal status, FIGO stage, or recurrence in patients with uterine cervical carcinoma, similar to the findings of several previous studies [27–29]. p53 overexpression is linked with the control of cell growth, cell cycle, and apoptosis [2,4,8–10]. Wild-type p53 protein usually resides in normal cell nuclei. This protein is unstable and has a half-life of only 20–30 min, while the mutant p53 protein is more stable and has a prolonged halflife, resulting in detectable immunohistochemical staining [8,10,21]. The mutation of the p53 gene is the most common genetic alteration found in many human carcinomas [2,4,8,9,30,36]. Epidemiological studies have shown that HPV infection, especially high-risk HPV including HPV types 16 and 18, is involved in the carcinogenesis of uterine cervical carcinoma [5–7]. It is known that high-risk HPV produces E6 oncoprotein, which binds to wild-type p53 protein to inactivate and degrade tumor suppression [3,9,10,16–20]. It is therefore postulated that loss of p53

Table 5 Clinicopathological parameters, p53 overexpression, HPV DNA, or disease free survival in recurrence cases Age

FIGO

Histology

Lymph node metastases

Parametrium involvement

p53 overexpression

HPV DNA

HPV Typing

Disease-free survival (months)

Prognoses

63 33 34 42 49 55 90 81 48 70 73

IA IB IB IB IB IB IB IIA IIB IIB IIB

SCC AC AC SCC SCC SCC SCC SCC AC SCC SCC

Unknown Unknown + + +
Unknown Unknown + + +


+
+ + +
+ + + +

+

+ +
+ + + + +

ND + ND + + ND +

+ +

ND 16 ND 16 16 ND 52 ND ND X 52

32 16 16 16 18 9 6 5 6 10 10

Survival Survival Survival Survival Death Death Death Survival Survival Survival Death

SCC: squamous cell carcinoma; AC: adenocarcinoma. ND: not done. X: unclassified HPV types.

534

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536

function either by binding to HPV coding proteins or by mutations of the p53 gene is an important event in the pathogenesis of cervical carcinomas [2,10,12,20]. The prevalence of p53 mutation varies among tumor types, ranging from 0 to 60% in major carcinomas [2,8– 10,15,30,37,38]. However, several studies have found that the p53 mutation is uncommon in uterine cervical carcinoma [10,36,39]. Immunohistochemistry has demonstrated elevated expression of p53 protein in a variety of human carcinoma, affecting the lung [40], prostate [15], bladder [15], ovary [11,41], endometrium [42–46], and uterine cervix [14,15,19,27–29,31–35,37,39,41–45,47]. Nuclear accumulation of p53 protein can be due to the presence of the mutated form, but stabilization of gene products other than point mutations and alterations in the normal degradation process or genetic alterations in genes other than the p53 gene can induce overexpression of p53 protein [48]. In addition, it is possible that p53 overexpression may be induced by wild-type p53 protein expression [48]. However, immunohistochemical studies have been widely carried out in many histopathological laboratories to screen for the presence of p53 protein, and it has been reported that p53 overexpression is related to progression or prognoses in many carcinomas [14,15,37,41–46]. In uterine cervical carcinoma, the detection rate of p53 overexpression by immunohistochemistry has been reported to range from 8 to 74% [20,27–35,39,47], and some reports have suggested that p53 overexpression is important as a prognostic marker [10,20,27–29,47]. On the other hand, there are few studies failing to demonstrate any prognostic value for p53 overexpression [16,30,31,33–35]. In uterine cervical carcinoma, prognostic factors such as clinical stage, lymph node metastases, or parametrium involvement have been established [22,23]. Our findings demonstrate a significant association between parametrium involvement and recurrence. Although FIGO stage and lymph node metastases showed slight associations with the frequency of recurrence, this difference was not significant. In this study, there was a strong correlation between p53 overexpression and FIGO stage, but there was no association between p53 overexpression and lymph node metastases. Although parametrium involvement is known as one of the most important prognostic factors in uterine cervical carcinoma, there have been very few reports on the association between p53 overexpression and parametrium invasion. Chen et al. [29] have reported that p53 overexpression shows no association with parametrium involvement. However, that study only investigated cervical carcinoma stage IB1. Huang et al. [27] have also demonstrated that p53 overexpression is not associated with parametrium invasion. In our study, p53 overexpression was observed more frequently in patients with parametrium involvement. There are several studies investigating whether HPV infection is associated with progression or prognoses in

uterine cervical carcinoma [49–55]. Though HPV infection does not correlate with clinical stage, the type of HPV DNA is associated with worse prognoses [49]. However, there is no relationship between HPV infection and prognoses [50– 52]. HPV infection shows a greater correlation with clinical features in the early stage compared with the advanced stage [53]. The absence of HPV DNA is associated with poorer prognoses than those in HPV-positive cervical carcinoma [54]. Multiple HPV infections are associated with shorter survival than that in patients with single HPV infection [55]. Our results suggest that the presence of HPV DNA is not correlated with age, clinical stage, or prognosis. However, there is slight association between HPV DNA and menopausal status. Several studies have implicated a relation between HPV infection and p53 status [9,10,18,20,36,56,57]. The existence of an inverse correlation between the presence of HPV DNA and p53 mutation has been reported in uterine cervical carcinoma [9,10,18,20,36,56,57]. In addition, several studies have shown that cervical carcinoma associated with HPV shows a wild-type p53 gene, while that without HPV shows a mutant p53 gene [9,10,18,20,36,56,57]. These results also suggest that the inactivation of p53 by mutation in the absence of HPV is an important step in carcinogenesis [18]. Immunohistochemical study would be expected to demonstrate an absence of p53 staining in HPV-positive cases and positive nuclear staining in HPV-negative cases with a missense mutation [10]. However, an other study has demonstrated that there is no inverse correlation between the degree of p53 immune activity and the ratio of HPV-positive cells [20]. Other studies have revealed no association between p53 overexpression and HPV DNA in cervical carcinoma [33,58], while it has been reported elsewhere that p53 overexpression is associated HPV infection [59]. Our study found no relationship between HPV infection and p53 overexpression, and further investigations with a large number of samples will be necessary to assess this phenomena. In the present study, either high-risk HPV infection or p53 overexpression was detected in 88% of all cases (22/25). Other investigators have reported similar findings [20]. This finding suggests that the inactivation of p53 function by HPV infection or altered p53 protein usually occurs in advance of cervical carcinoma [20]. The results of the present study indicate that p53 overexpression is more frequent in older than in younger patients, suggesting that there are some specific genetic events influencing p53 overexpression in older patients. Although it has previously been reported that HPV positivity is related to patient age [48,60], there was no association observed between HPV DNA and age in the present study. In conclusion, though the observation period was brief in the present study, investigating p53 overexpression by immunohistochemical study may be useful in identifying cases at risk for recurrence in uterine cervical carcinoma. We

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536

examined a small number of cases in the present study, and other studies should be performed with a more population and a long-term follow-up to elucidate p53 overexpression by immunohistochemical study as a prognostic factor for uterine cervical carcinoma.

[18]

[19] [20]

Acknowledgements [21]

The authors would like to thank Drs. Maki Tanaka and Yoshiaki Tanaka (Department of gynecology, TerakataIkuno Hospital, Osaka, Japan) for their advice. Secretarial assistance from Ms. Ayuko Asaoka, Ms. Wakako Okamoto, Ms. Miyuki Imai, and Ms. Noriko Sugie is also greatly appreciated.

[22]

[23]

[24]

References [25] [1] Pisani P, Bray F, Parkin DM. Estimates of the world-wide prevalence of cancer for 25 sites in the adult population. Int J Cancer 2002;97:72– 81. [2] Levine AJ, Perry ME, Chang A, et al. The 1993 Walter Hubert Lecture: the role of the p53 tumor-suppressor gene in tumorigenesis. Br J Cancer 1994;69:409–16. [3] Tommasino M, Accardi R, Caldeira S, et al. The role of TP53 in cervical carcinogenesis. Hum Mutat 2003;21:307–12. [4] Hollstein M, Sidransky D, Vogelstein B, et al. p53 Mutations in human cancers. Science 1991;253:49–53. [5] Walboomers JMM. Jacobs MV, Manos MM, et al. Human papilloma virus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;189:12–9. [6] zur Hauzen H. Papilloma viruses and cancer: from basic studies to clinical application. Nat Rev Cancer 2002;2:342–50. [7] Mun˜oz N. Human papilloma virus and cancer: the epidemiological evidence. J Clin Virol 2000;19:1–5. [8] Levine AJ. The p53 tumor-suppressor gene. N Eng J Med 1992; 14:1350–2. [9] Levine AJ, Momand J, Finlay CA. The p53 tumour suppressor gene. Nature 1991;351:453–6. [10] Greenblatt MS, Bennett WP, Hollstein M, et al. Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res 1994;54:4855–78. [11] Kohler MF, Kerns BJM, Humphrey PA, et al. Mutation and overexpression of p53 in early-stage epithelial ovarian cancer. Obstet Gynecol 1993;81:643–50. [12] Livingstone LR, White A, Sprouse J, et al. Altered cell cycle arrest and gene amplification potential accompany loss of wild-type p53. Cell 1992;70:923–35. [13] Ahn WS, Han YJ, Bae SM, et al. Differential suppression of human cervical cancer cell growth by adenovirus delivery of p53 in vitro: arrest phase of cell cycle is dependent on cell line. Jpn J Caner Res 2002;93:1012–9. [14] Verstegen MJT, Troost D, Leenstra S, et al. Proliferation- and apoptosis-related proteins in intracranial ependymomas: an immunohistochemical analysis. J Neuro-oncol 2002;56:21–8. [15] Algaba F, Trias I, Santinelli A, et al. TP53 in urologic tumors. AQCH 2003;25:123–30. [16] Graflund M, Sorbe B, Karlsson M. Immunohistochemical expression of p53, bcl-2, and p21WAF1/CIP1 in early cervical carcinoma: correlation with clinical outcome. Int J Gynecol Cancer 2002;12:290–8. [17] Kedzia W, Schmidt M, Frankowski A, et al. Immunohistochemical assay of p53, cyclin D1, c-erbB2, EGFr and Ki-67 proteins in HPV-

[26]

[27]

[28]

[29]

[30] [31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

535

positive and HPV-negative cervical cancers. Folia Histochem Cytobiol 2002;40:37–41. Crook T, Wrede D, Tidy JA, et al. Clonal p53 mutation in primary cervical cancer: association with human papilloma virus-negative tumors. Lancet 1992;339:1070–3. Werness BA, Levine AJ, Howley PM. Association of human papilloma virus types 16 and 18 E6 proteins with p53. Science 1990;248:76–9. Akasofu M, Oda Y. Immunohistochemical detection of p53 in cervical epithelial lesions with or without infection of human papilloma virus types 16 and 18. Virchow Archiv 1995;425:593–602. Oka M, Kounoura K, Narasaki F, et al. P-glycoprotein is positively correlated with p53 protein accumulation in human colorectal cancers. Jpn J Cancer Res 1997;88:738–42. Perez CA, Kurman RJ, Stehman FB, Thipgen JT. Uterine cervix. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. Philadelhpia: Lippincott; 1993. p. 591–662. Fu YS, Reagan JW. Pathology of the uterine cervix, vagina, and vulva. In: Bennington JL, editor. Major problems in pathology. Philadelphia, PA: WB Saunders; 1989. p. 269–77. Saiki RK, Scharf S, Faloona F, et al. Enzymatic amplification of bglobin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 1985;230:1350–4. Roda Husman AM, Walboomers JM, Meijer CJ, et al. Analysis of cytomorphologically abnormal cervical scrapes for the presence of 27 mucosotropic human papilloma virus genotypes, using polymerase chain reaction. Int J Cancer 1994;56:802–6. Yoshikawa H, Kawana T, Kitagawa K, et al. Detection and typing of multiple genital human papilloma viruses by DNA amplification with consensus primers. Jpn J Cancer Res 1991;82:524–31. Huang LW, Chou YY, Chao SL, et al. p53 and p21 expression in precancerous lesions and carcinomas of the uterine cervix: overexpression of p53 predicts poor disease outcome. Gynecol Oncol 2001;83:348–54. ˚ vall-Lundqvist EH, Silfverswa¨rd C, Aspenblad U, et al. The impact A of tumour angiogenesis, p53 overexpression and proliferative activity (MIB-1) on survival in squamous cervical carcinoma. Eur J Cancer 1997;33:1799–804. Chen HY, Hsu CT, Lin WC, et al. Prognostic value of p53 expression in stage IB1 cervical carcinoma. Gynecol Obstet Invest 2000;49:266– 71. Oka K, Nakano T, Arai T. P53CM1 expression is not associated with prognosis in uterine cervical carcinoma. Cancer 1993;72:160–4. Brenna SMF, Zeferino LC, Pinto GA, et al. p53 expression as a predictor of recurrence in cervical squamous cell carcinoma. Int J Gynecol Cancer 2002;12:299–303. Helland A, Holm R, Kristensen G, et al. Genetic alterations of the TP53 gene, p53 protein expression and HPV infection in primary cervical carcinomas. J Pathol 1993;171:105–14. Cavuslu S, Goodlad J, Hobbs C, et al. Relationship between human papilloma virus infection and overexpression of p53 protein in cervical carcinomas and lymph node metastases. J Med Virol 1997;53:111–7. Dejonge ETM, Viljoen E, Lindeque BG, et al. The prognostic significance of p53, mdm2, c-erbB-2, cathepsin D, and thrombocytosis in stage IB cervical cancer treated by primary radical hysterectomy. Int J Gynecol Cancer 1999;9:198–205. ter Harmsel B, van Muyden R, Smedts F, et al. The significance of cell type and tumor growth markers in the prognosis of unscreened cervical cancer patients. Int J Gynecol Cancer 1998;8:336–44. Tsuda H, Hirohashi S. Frequent occurrence of p53 gene mutations in uterine cancers at advanced clinical stage with aggressive histological phenotypes. Jpn J Cancer Res 1992;83:1184–91. Sisk EA, Soltys SG, Zhu S, et al. Human papilloma virus and p53 mutational status as prognostic factors in head and neck carcinoma. Head Neck 2002;24:841–9. Kihana T, Tsuda H, Teshima S, et al. High incidence of p53 gene mutation in human ovarian cancer and its association with nuclear

536

[39] [40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

A. Ikuta et al. / Cancer Detection and Prevention 29 (2005) 528–536 accumulation of p53 protein and tumor DNA aneuploidy. Jpn J Cancer Res 1992;83:978–84. Denk C, Butz K, Schneider A, et al. p53 Mutations are rare events in recurrent cervical cancer. J Mol Med 2001;79:283–8. Simada H, Okazumi S, Takeda A, et al. Presence of serum p53 antibodies is associated with decreased in vitro chemosensitivity in patients with esophageal cancer. Surg Today 2001;31:591–6. Halperin R, Zehavi S, Dar P, et al. Clinical and molecular comparison between borderline serous ovarian tumors and advanced serous papillary ovarian carcinomas. Eur J Gynecol Oncol 2001;22:292–6. Terlikowski S, Lenczewski A, Famulski W, et al. Patterns of immunohistochemical staining for p53 expression in hyperplastic endometrium and adenocarcinoma. Folia Histochem Cytobiol 2001;39:195–6. Powell B, Soong R, Grieu F, et al. p53 Protein overexpression is a prognostic indicator of poor survival in stage I endometrial carcinoma. Int J Oncol 1999;14:175–9. Salvesen HB, Iversen OE, Akslen LA. Prognostic significance of angiogenesis and Ki-67, p53, and p21 expression: a population-based endometrial carcinoma study. J Clinic Oncol 1999;17:1382–90. Ohkouchi T, Sakuragi N, Watari H, et al. Prognostic significance of Bcl-2, p53 overexpression, and lymph node metastasis in surgically staged endometrial carcinoma. Am J Obstet Gynecol 2002;187:353–9. Sung CJ, Zheng Y, Quddus MR, et al. p53 As a significant prognostic marker in endometrial carcinoma. Int J Gynecol Cancer 2000;10: 119–27. ˚ , et al. Immunohistochemical analysis Holm R, Skomedal H, Helland A of p53 protein overexpression in normal, premalignant, and malignant tissues of the cervix utreri. J Pathol 1993;169:21–6. Saito J, Hoshiai H, Noda K. Type of human papilloma virus and expression of p53 in elderly women with cervical cancer. Gynecol Obstet Invest 2000;49:190–3. Nakagawa S, Yoshikawa H, Onda T, et al. Type of human papilloma virus is related to clinical features of cervical carcinoma. Cancer 1996; 78:1935–41.

[50] Kristensen GB, Karlsen F, Jenkins A, et al. Human papilloma virus has no prognostic significance in cervical carcinoma. Eur J Cancer 1996; 32A:1349–53. [51] van Muyden RCPA, ter Harmsel BWA, Smedts FMM, et al. Detection and typing of human papilloma virus in cervical carcinomas in Russian women: a prognostic study. Cancer 1999;85:2011–6. [52] King LA, Tase T, Twiggs LB, et al. Prognostic significance of the presence of human papilloma virus DNA in patients with invasive carcinoma of the cervix. Cancer 1989;63:897–900. [53] Schwartz SM, Daling JR, Shera KA, et al. Human papilloma virus and prognosis of invasive cervical cancer: a population-based study. J Clin Oncol 2001;19:1906–15. [54] Riou G, Favre M, Jeannel D, et al. Association between poor prognosis in early-stage invasive cervical carcinomas and non-detection of HPV DNA. Lancet 1990;335:1171–4. [55] Bachtiary B, Obermair A, Dreier B, et al. Impact of multiple HPV infection on response to treatment and survival in patients receiving radical radiotherapy for cervical cancer. Int J Cancer 2002;102: 237–43. [56] Sobti RC, Parashar K, Kaur R, et al. Detection of human papilloma virus DNA, serum p53, and p53 antibodies in patients with cervical cancer. JEPTO 2002;21:79–85. [57] Basari S, Roncalli M, Viale G, et al. p53 immunoreactivity in inflammatory and neoplastic disease of the uterine cervix. J Pathol 1993;169:425–30. [58] Bar JK, Harlozinska A, Sedlaczek P, et al. Relations between the expression of p53, c-erbB-2, Ki-67 and HPV infection in cervical carcinomas and cervical dysplasias. Anticancer Res 2001;21:1001–6. [59] Ngan HYS, Cheung ANY, Liu SS, et al. Abnormal expression of panras, c-myc and tp53 in squamous cell carcinoma of cervix: correlation with HPV and prognosis. Oncol Rep 2001;8:557–61. [60] Yang YC, Shen J, Tage JE, et al. Cervical cancer in young in Taiwan: prognosis is independent of papilloma virus or tumor cell type. Gynecol Oncol 1997;64:59–63.