Annals of Oncology 14: 214219, 2003

DOI: 10.1093/annonc/mdg071

Original article

Chemosensitivity and p53Bax pathway-mediated apoptosis in

patients with uterine cervical cancer
H. Sultana1, J. Kigawa1*, Y. Kanamori1, H. Itamochi1, T. Oishi1, S. Sato1, S. Kamazawa1, M. Ohwada2,
M. Suzuki2 & N. Terakawa1
1

Department of Obstetrics and Gynecology, Tottori University School of Medicine, Yonago; 2Department of Obstetrics and Gynecology, Jichi Medical School Hospital,
Ustunomiya, Japan
Received 21 May 2002; revised 26 July 2002; accepted 17 October 2002

Objectives: To determine whether and how apoptosis through the p53Bax pathway affects sensitivity to
chemotherapy in cervical cancer.

Materials and methods: Thirty patients with cervical squamous cell carcinoma, who had human papilloma
virus (HPV) and underwent neoadjuvant chemotherapy, were entered in the present study. Tumor specimens
were obtained before and after chemotherapy. HPV was detected by polymerase chain reaction. The expression of Ki-67, p53, Bax and Bcl-2 proteins was determined by immunohistochemical staining. Apoptotic cells
were identified by the terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate biotin
nick-end labeling method.
Results: Of 30 patients, 18 responded to chemotherapy and 12 did not. The apoptotic index in tumors of
responders was significantly higher than in non-responders after chemotherapy. The Ki-67 labeling index (LI)
in responders was significantly higher than in non-responders before chemotherapy. Patients with tumors
>33% of the LI, which was determined by a receiver operating characteristic curve, had a better survival rate.
The incidence of p53 protein expression did not differ between responders and non-responders. After chemotherapy, the expression of Bax protein in responders was more frequent and Bcl-2 protein expression was less
frequent than in non-responders.
Conclusions: Chemosensitivity in cervical cancer may be associated with apoptosis via the p53Bax
pathway.
Key words: apoptosis, Bax, cervical carcinoma, chemotherapy, p53, uterus

Introduction
Platinum-based chemotherapy has recently proved to be an
effective therapy for cervical cancer [1, 2]. Several studies have
shown that chemoradiation improved survival rate in patients
with cervical cancer [3, 4]. Previously published work also suggests that intra-arterial chemotherapy may improve the prognosis
of patients with advanced cervical cancer [5, 6], but is limited by
chemoresistance. Many mechanisms have been postulated to
explain chemoresistance, including decreased drug accumulation
inside tumor cells, increased cellular detoxification, and increased
DNA repair activity [79]. Recent studies indicate that apoptosis
also contributes to chemosensitivity [10, 11].
p53 is known to be a cell cycle checkpoint protein playing a
regulatory role in the control of cell proliferation and apoptosis
[12]. Although cervical squamous cell carcinoma commonly
contains a wild-type p53 gene, it is highly correlated with human
papilloma virus (HPV) infection. Because the viral oncoprotein
*Correspondence to: Dr J. Kigawa, Department of Obstetrics and
Gynecology, Tottori University School of Medicine, 36-1 Nisimachi
Yonago, 6838504 Japan. Tel: +81-0859-34-8127; Fax: +81-0859-34-8089;
E-mail: kigawa@grape.med.tottori-u.ac.jp
2003 European Society for Medical Oncology

E6 binds and inhibits the function of p53 protein, inhibition by

HPV may be one cause of chemoresistance in cervical cancer
[13, 14]. However, in cervical cancer the relationship between
apoptosis through the p53 pathway and chemosensitivity is not
clear.
Ki-67, a marker for cellular proliferation, has been applied to
study the growth fraction and cell-kinetic activities [15]. Bax,
which is regulated by the p53 gene, controls apoptosis, and Bcl-2
opposes Bax function [1618]. We conducted the present study
to determine whether and how apoptosis through the p53Bax
pathway affects sensitivity to chemotherapy in cervical cancer.

Materials and methods

Patients
A total of 32 patients with cervical squamous cell carcinoma, who received
primary cisplatin-based neoadjuvant chemotherapy followed by radical
hysterectomy between 1990 and 2000 at Tottori University Hospital and Jichi
Medical School Hospital, were entered in the present study. According to
the International Federation of Gynecology and Obstetrics (FIGO) staging
system, six cases were stage IB, 18 cases were stage II (three stage IIA and
15 stage IIB ), and eight cases were stage IIIA. The mean age for patients was

215
49.2 years (range 2466 years). Informed consent was obtained from all
patients.
According to Minagawas protocol [6], all patients received bleomycin
3.5 mg/m2 i.v. on days 15; vincristin 0.7 mg/m2 and mitomycin C 10 mg i.v.
on day 5; and 25 mg/m2 cisplatin infused via each internal iliac artery on
day 5. Cisplatin was injected for 20 min into both of the internal iliac arteries
according to the Seldinger [19] technique.
Twenty-nine patients underwent two cycles whilst the remaining three
patients received three cycles. Three weeks after each course of chemotherapy, we evaluated therapeutic efficacy according to the following criteria
using both computed tomography and magnetic resonance imaging for all
patients. Complete response (CR) was defined as absence of disease; partial
response (PR) was defined as a >50% reduction in all measurable lesions
without the appearance of new lesions; no change (NC) was defined as a
<50% decrease or a <25% increase in all measurable lesions without the
appearance of new lesions. Progressive disease was defined as a >25%
increase in the measurable disease at a known site or the appearance of new
lesions. To obtain a specimen before chemotherapy, a biopsy was carried out
under colposcopy. A surgical specimen was used after chemotherapy.

HPV detection
Human papilloma virus DNA was examined by the polymerase chain
reaction (PCR). Genomic DNA was extracted from the paraffin-embedded
tissue of the most severe lesion of the surgical specimen. In brief, tissue
blocks were cut into 10 m sections using a microtome. Five tissue sections
were deparaffinized twice in xylene, hydrated in graded alcohols, and incubated in proteinase K buffer consisting of 10 mM TrisHCl, 10 mM EDTA,
proteinase K 100 g/ml and 0.5% sodium dodecyl sulfate at 37C, overnight.
The supernatant fluid was treated twice with phenol and once with chloroform. A precipitate of DNA was obtained by adding 3 M sodium acetate and
100% ethanol. The precipitate was washed in 70% ethanol and dissolved in
distilled water. -globin primer was used as an internal control for the suitability of DNA in the samples. We used a pair of consensus primers with
the ability to detect HPV types 6, 11, 16, 18, 31, 33, 42, 52 and 58 [20]. The
DNA sequences of the consensus primer pairs were 5-CGTAAACGTTTTCCCTATTTTTTT-3 and 5-TACCCTAAATACTCTGTATTG-3. Human
papilloma virus DNA was amplified in 50 l of a reaction mixture containing
0.5 g of sample DNA, 50 mM potassium chloride, 10 mM TrisHCl
(pH 8.8), 1.5 mM magnesium chloride, 0.1% Triton X-100, 200 M of deoxyribonucleoside triphosphate, 0.5 M of each primer, and 1 U of Taq DNA
polymerase (Wako, Osaka, Japan). The sample was subjected to 35 cycles of
amplification on a PCR processor (PC-700; Astec, Fukuoka, Japan). Each
cycle consisted of DNA denaturing for 1.5 min at 95C, annealing for 1.5 min
at 48C, and extension for 2 min at 72C. An aliquot of 10 l of the reaction
mixture was electrophoresed on 4% agarose gel with ethidium bromide
staining. As a result, HPV DNA was detected in 30 of 32 patients (93.7%).
We examined 30 patients with HPV-positive tumor in a further study.

(1:100 dilution; Calbiochem, San Diego, LA), respectively. The primary antibody was visualized using the Histofine Simple Stain PO kit (Nichirei,
Tokyo, Japan) according to the instruction manual. For the negative controls,
the primary antibodies were replaced with phosphate-buffered saline. The
slides were counterstained with hematoxylin.
For Ki-67 staining slides, the labeled and unlabeled cells were counted
in five high-power fields (400). A total of 500 cells were counted in each
specimen. The Ki-67 labeling index (LI) was determined using the following
formula: LI (%) = 100 labeled cells/total cells. For expression of p53, Bax
and Bcl-2 proteins, a positive case was defined as staining of the tumor cells
and a negative case was defined as no staining of any tumor cells.

Detection of apoptosis
Apoptotic cells were identified by the terminal deoxynucleotidyl transferasemediated deoxyuridine triphosphate biotin nick-end labeling (TUNEL)
method using the Apop Taq in situ detection kit (Oncor Inc., Gaithersburg,
MD). Dewaxed and rehydrated specimens were incubated in proteinase K
40 g/ml for 1 h at 37C and were treated with 3% H2O2 in methanol for
30 min at room temperature. After adding equilibration buffer for 5 min at
room temperature, terminal deoxynucleotidyl transferase (TdT) enzyme was
pipetted onto the sections and incubated at 37C for 2 h. The reaction was
stopped by incubating the sections in stop buffer for 30 min at 37C. Antidigoxigenin peroxidase was added to the slides, followed by incubation for
30 min at 37C. Slides were stained with diaminobenzine for 10 min and
counterstained with hematoxylin. A total of 500 cells were counted in each
specimen. The apoptotic index (AI) was defined as follows: AI (%) = 100
apoptotic cells/total cells.

Statistical analysis
Patient survival distribution was calculated using the KaplanMeier
method. The significance of the survival distribution in each group was tested
by a log-rank test. Values are expressed as mean standard deviation (SD).
Statistical analysis was performed using the Stat View Version 5.0-J program
(Hulinks Inc., Tokyo Japan). A value of P <0.05 was considered statistically
significant.

Results
Of the 30 patients, 18 responded to chemotherapy (PR) and 12
did not (NC). Responders were significantly younger than non-

Table 1. Patient characteristics

Characteristic

Non-responders (n = 12)

3366

2462

53.3

43.1

Age (years)
Range

Immunohistochemistry
Immunohistochemical studies were performed to detect Ki-67, p53, Bax and
Bcl-2 proteins. Formalin-fixed, paraffin-embedded tissue sections were
mounted on silane-coated glass slides. Deparaffinized and rehydrated
samples were heated in a microwave oven for 15 min at 94C in citrate buffer
solution. The slides were cooled and endogeneous peroxidase was blocked
with 3% hydrogen peroxide (H2O2) in methanol for 20 min at room temperature, followed by rinsing in distilled water. To detect Ki-67, p53, Bax and
Bcl-2 proteins, the sections were incubated overnight at 4C with anti-Ki-67
monoclonal antibody, MIB-1 (1:50 dilution; Immunotech, Marseille, France),
anti-p53 monoclonal antibody, DO-7 (1:50 dilution; DAKO HS, Glostrup,
Denmark), anti-Bax polyclonal antibody, N-20 (1:200 dilution; Santa Cruz
Biotechnology, Santa Cruz, CA) and anti-bcl-2 monoclonal antibody, Ab-3

Responders (n = 18)

Mean

Histology
Keratinizing

16

10

II

10

III

Non-keratinizing
FIGO stage

P = 0.012.
FIGO, International Federation of Gynecology and Obstetrics.

216
responders (P = 0.012). There were no differences in histological
grade and FIGO stage between the two groups (Table 1). The
5-year survival rate for responders was significantly better than
for non-responders (92.9% versus 46.9%, P <0.001).
Figure 1 shows a representative case of TUNEL and immunohistochemical stainings for Ki-67, Bax and Bcl-2 proteins. The
AI did not differ between responders and non-responders before
chemotherapy. After chemotherapy, the AI was significantly
increased in both groups (P = 0.03; Table 2), and was significantly
higher for responders than for non-responders (P = 0.018). Before
chemotherapy, the Ki-67 LI in the tumors of responders was
significantly higher compared with non-responders (P <0.001).
The Ki-67 LI for responders was significantly decreased after
chemotherapy (P <0.01), but that for non-responders was not.

The receiver operating characteristic curve demonstrated that the

cut-off value of Ki-67 LI before chemotherapy was 33.0% [21].
The response rate for patients with tumors >33.0% LI was significantly higher than for those with tumors <33.0% LI (94.1%
versus 15.4%). Patients with tumors >33.0% LI showed a better
survival rate (Figure 2).
The incidence of p53 expression did not differ between
responders and non-responders either before or after chemotherapy. Neither Bax nor Bcl-2 expression differed between
responders and non-responders before chemotherapy. After, but
not before, chemotherapy, the expression of Bax protein was
observed more frequently and Bcl-2 protein expression was
observed less frequently in responders, compared with nonresponders (Table 3).

Figure 1. TUNEL and immunohistochemical stainings in a patient after chemotherapy. (A) TUNEL, (B) Ki-67, (C) Bax (positive), and (D) Bcl-2 (positive).

217
Table 2. Apoptotic index and Ki-67 labeling index before and after chemotherapy
Response to chemotherapy

Apoptotic index

Ki-67 labeling index

Before chemotherapy

After chemotherapy

Before chemotherapy

After chemotherapy

Responders

0.8 2.0

2.3 2.8

41.7 7.2

34.3 10.9d

Non-responders

0.1 0.1

0.5 0.7

29.0 8.3

29.1 6.6

a
e

c
g

a
versus b and e versus f, P = 0.03; b versus f, P = 0.018; c versus g, P <0.001; c versus d, P <0.01 (Wilcoxon signed-rank test and
MannWhitney U-test).

Table 3. The expression of Bax and Bcl-2 before and after chemotherapy
Response to chemotherapy

Bax
Before chemotherapy (%)

Responders (n = 18)
Non-responders (n = 12)
a

8 (44.4)
2 (16.6)

Bcl-2
After chemotherapy (%)

Before chemotherapy (%)

After chemotherapy (%)

10 (55.6)

9 (50.0)

5 (27.8)b

2 (16.6)

6 (50.0)

8 (66.7)b

P = 0.038.
P = 0.042.

Discussion
Several authors examined the prognostic value of tumor proliferation and apoptosis in patients with carcinoma of the uterine
cervix, but the results have been controversial [2227]. In
patients with cervical cancer undergoing radiotherapy, the Ki-67
LI was a significant factor for disease-free survival, but the AI
was not [22]. One author showed that the Ki-67 LI was not related
to outcome of radiotherapy; however, a high AI proved to be an
indicator for poor outcome [23, 24]. On the other hand, in patients
treated with primary surgery, neither the AI nor the Ki-67 LI had
prognostic significance [24]. The differential prognostic relevance of the AI for cervical cancer treated with radiotherapy or
surgery may be explained by the evidence that tumors with a high
AI are hypoxic. Hypoxia is known to be associated with radiation
ineffectiveness [28]. In the present study with chemotherapy, the

Figure 2. Estimated survival rate. With the cut-off value of Ki-67 LI before
chemotherapy set at 33.0%, patients with a >33.0% LI had a better survival
rate.

AI did not differ between responders and non-responders before

treatment. In contrast, before chemotherapy, the Ki-67 LI for
responders was significantly higher than for non-responders. It is
known that rapidly proliferating cells are more sensitive to cytotoxic agents than slowly proliferating cells [29, 30]. In another
study, intracellular drug accumulation decreased in resting cells
[31]. Accordingly, the Ki-67 LI may be a parameter for chemoresponse and prognosis in cervical cancer.
The aim of the present study was to determine whether and
how apoptosis through the p53Bax pathway affects sensitivity
to chemotherapy in cervical cancer. We examined cervical cancer
patients with HPV-positive tumor. Because the present study had
no control group, no real comment could be made. Human
papilloma virus is detected in >90% of cervical cancers and E6
protein, encoded by HPV, inhibits the function of p53 protein.
This suggests that HPV may be directly related to pathways
regulating apoptosis [32].
We failed to find a correlation between p53 protein expression
by immunohistochemical staining and chemotherapy-induced
apoptosis. Previously, overexpression of the p53 protein was
believed to be caused by an underlying abnormal p53 gene, leading to the expression of an abnormal and stabilized protein [33].
In the literature, the expression of p53 protein did not affect
apoptosis [34]. These findings suggest that immunohistochemical
staining is not an appropriate technique to assess p53 function. In
addition, p53 expression may be suppressed by HPV in cervical
cancer.
To determine p53Bax pathway-mediated apoptosis, we examined the expression of Bax and Bcl-2 proteins. p53 is a direct
transcriptional activator of the Bax gene, but Bcl-2 blocks both
p53-dependent and p53-independent pathways. In the literature,
the proportion of Bax-positive cells was significantly higher in
responders with advanced cervical cancer treated by cisplatinbased chemotherapy, but no significant difference was found in
Bcl-2 protein expression between responders and non-responders

218
[35]. In contrast, Tjalma et al. showed a strong relationship
between Bcl-2 protein expression and overall survival in cervical
cancer [36]. However, it is noteworthy that these findings were
seen only before treatment. Interestingly, the present study showed
that the expression of Bax protein was observed more frequently
and Bcl-2 protein expression less frequently in responders after
chemotherapy. Harima et al., investigating the expression of Bax
and Bcl-2 proteins before and during the course of radiation
therapy in cervical cancer, found that better tumor control was
accompanied by increased Bax protein expression [26]. An
in vitro study showed that expression of p53 and Bax proteins
increased and expression of Bcl-2 protein decreased after exposure
to DNA-damaging agents [37].
In conclusion, the present study suggests that apoptosis via the
p53Bax pathway is associated with response to cisplatin-based
chemotherapy in cervical cancer.

References
1. Panici PB, Greggi S, Scambia G et al. High-dose cisplatin and bleomycin
neoadjuvant chemotherapy plus radical surgery in locally advanced
cervical carcinoma: preliminary report. Gynecol Oncol 1991; 41:
212216.

13. Villa LL. Human papilloma viruses and cervical cancer. Adv Cancer Res
1997; 71: 321341.
14. Scheffner M. Ubiquitin, E6-AP, and their role in p53 inactivation.
Pharmacol Ther 1998; 78: 129139.
15. Mochen C, Giardini R, Costa A, Silvestrini R. MIB-1 and S-phase cell
fraction predict survival in non-Hodgkins lymphomas. Cell Prolif 1997;
30: 3747.
16. Miyashita T, Reed JC. Tumor suppressor p53 is a direct transcriptional
activator of the human bax gene. Cell 1995; 80: 293299.
17. Oltvai ZN, Milliman CL, Korsmeyer SJ. Bcl-2 heterodimerizes in vivo
with a conserved homolog, Bax, that accelerates programmed cell death.
Cell 1993; 74: 609619.
18. Boise LH, Gonzalez-Garcia M, Postema CE et al. Bcl-x, a bcl-2-related
gene that functions as a dominant regulator of apoptotic cell death. Cell
1993; 74: 597608.
19. Seldinger SI. Catheter replacement of the needle in percutaneous arteriography. Acta Radiol 1953; 39: 368376.
20. Yoshikawa H, Kawana T, Kitagawa K et al. Detection and typing of
multiple genital human papillomaviruses by DNA amplification with
consensus primers. Jpn J Cancer Res 1991; 82: 524531.
21. Metz CE. Basic principles of ROC analysis. Semin Nucl Med 1978; 8:
486498.
22. Tsang RW, Wong CS, Fyles AW et al. Tumour proliferation and
apoptosis in human uterine cervix carcinoma II: correlations with clinical
outcome. Radiother Oncol 1999; 50: 93101.

2. Sardi J, Sananes C, Giaroli A et al. Neoadjuvant chemotherapy in locally

advanced carcinoma of the cervix uteri. Gynecol Oncol 1990; 38:
486493.

23. Levine EL, Renehan A, Gossiel R et al. Apoptosis, intrinsic radiosensitivity and prediction of radiotherapy response in cervical carcinoma.
Radiother Oncol 1995; 37: 19.

3. Keys HM, Bundy BN, Stehman FB et al. Cisplatin, radiation, and adjuvant hysterectomy compared with radiation and adjuvant hysterectomy
for bulky stage IB cervical carcinoma. N Engl J Med 1999; 340: 1154
1161.

24. Levine EL, Davidson SE, Roberts SA et al. Apoptosis as predictor of

response to radiotherapy in cervical carcinoma. Lancet 1994; 344: 472.

4. Thomas GM. Improved treatment for cervical cancerconcurrent

chemotherapy and radiotherapy. N Engl J Med 1999; 340: 11981200.
5. Kigawa J, Minagawa Y, Ishihara H et al. The role of neoadjuvant intraarterial infusion chemotherapy with cisplatin and bleomycin for locally
advanced cervical cancer. Am J Clin Oncol 1996; 19: 255259.

25. Hockel M, Schlenger K, Hockel S, Vaupel P. Hypoxic cervical cancers

with low apoptotic index are highly aggressive. Cancer Res 1999; 59:
45254528.
26. Harima Y, Nagata K, Harima K et al. Bax and Bcl-2 protein expression
following radiation therapy versus radiation plus thermoradiotherapy in
stage IIIB cervical carcinoma. Cancer 2000; 88: 132138.

6. Minagawa Y, Kigawa J, Irie T et al. Radical surgery following neoadjuvant chemotherapy for patients with stage IIIB cervical cancer. Ann
Surg Oncol 1998; 5: 539543.

27. Yuki H, Fujimura M, Yamakawa Y et al. Detection of apoptosis and

expression of apoptosis-associated proteins as early predictors of
prognosis after irradiation therapy in stage IIIb uterine cervical cancer.
Jpn J Cancer Res 2000; 91: 127134.

7. Andrews PA, Velury S, Mann SC, Howell SB. cis-Diaminedichloroplatinum (II) accumulation in sensitive and resistant human ovarian
carcinoma cells. Cancer Res 1988; 48: 6873.

28. Gray LH, Conger AD, Ebert M et al. The concentration of oxygen
dissolved in tissues at the time of irradiation as a factor in radiotherapy.
Br J Radiol 1953; 26: 638648.

8. Lautier D, Canitrot Y, Deeley RG, Cole SP. Multidrug resistance

mediated by the multidrug resistance protein (MRP) gene. Biochem
Pharmacol 1996; 52: 967977.

29. Bonetti A, Zaninelli M, Rodella S et al. Tumor proliferative activity and

response to first-line chemotherapy in advanced breast carcinoma. Breast
Cancer Res Treat 1996; 383: 289297.

9. Kigawa J, Minagawa Y, Cheng X, Terakawa N. Gamma-glutamyl

cysteine synthetase up-regulates glutathione and multidrug resistanceassociated protein in patients with chemoresistant epithelial ovarian
cancer. Clin Cancer Res 1998; 4: 17371741.

30. OReilly SM, Camplejohn RS, Rubens RD, Richards MA. DNA flow
cytometry and response to preoperative chemotherapy for primary breast
cancer. Eur J Cancer 1992; 28: 681683.

10. Righetti SC, Della Torre G, Pilotti S et al. A comparative study of p53
gene mutations, protein accumulation, and response to cisplatin-based
chemotherapy in advanced ovarian carcinoma. Cancer Res 1996; 56:
689693.
11. Minagawa Y, Kigawa J, Itamochi H et al. Cisplatin-resistant HeLa cells
are resistant to apoptosis via p53-dependent and -independent pathways.
Jpn J Cancer Res 1999; 90: 19731979.
12. Williams GT, Smith CA. Molecular regulation of apoptosis: genetic
controls on cell death. Cell 1993; 74: 777779.

31. Dimanche-Boitrel MT, Pelletier H, Genne P et al. Confluence-dependent

resistance in human colon cancer cells: role of reduced drug accumulation and low intrinsic chemosensitivity of resting cells. Int J Cancer
1992; 50: 677682.
32. Shoji Y, Saegusa M, Takano Y et al. Correlation of apoptosis with
tumour cell differentiation, progression, and HPV infection in cervical
carcinoma. J Clin Pathol 1996; 49: 134138.
33. Kraiss S, Spiess S, Reihsaus E, Montenarh M. Correlation of metabolic
stability and altered quanternacy structure of oncoprotein p53 with cell
transformation. Exp Cell Res 1991; 192: 157164.

219
34. Sato S, Kigawa J, Minagawa Y et al. Chemosensitivity and p53dependent apoptosis in epithelial ovarian carcinoma. Cancer 1999; 86:
13071313.
35. Kokawa K, Shikone T, Otani T, Nakano R. Apoptosis and the expression
of Bax and Bcl-2 in squamous cell carcinoma and adenocarcinoma of the
uterine cervix. Cancer 1999; 85: 17991809.

36. Tjalma W, De Cuyper E, Weyler J et al. Expression of bcl-2 in invasive

and in situ carcinoma of the uterine cervix. Am J Obstet Gynecol 1998;
178: 113117.
37. Zhan Q, Bieszczad CK, Bae I et al. Induction of BCL-2 family member
MCL1 as an early response to DNA damage. Oncogene 1997; 14: 1031
1039.