Anatomic Pathology / FASL AND BCL-2 IN CERVICAL CARCINOMA
Expression of Fas Ligand and bcl-2 in Cervical Carcinoma
and Their Prognostic Significance
Satoru Munakata, MD, PhD,1 Omi Watanabe, MD,2 Kazutomo Ohashi, MD, PhD,2
and Hideo Morino, MD, PhD1
Key Words: Cervical carcinoma; Immunohistochemistry; Fas ligand; bcl-2; Survival
DOI: 10.1309/0773N4Q3GFP34J5V
Abstract
Apoptosis-related factors are known to influence
survival with many malignant tumors. We performed
immunohistochemical analysis of Fas ligand (FasL) and
bcl-2 in invasive cervical cancer to determine the
association with prognosis. In 125 patients with
cervical cancer, 93 cases (74.4%) were positive for
FasL, and 94 cases (75.2%) were positive for bcl-2.
When 101 cases, clinical stages I through IV, were
analyzed by univariate analysis, negative bcl-2 (P =
.035) and combined positive FasL and negative bcl-2
(PFNB; P = .0025) were associated with significantly
decreased disease-free survival. Positive FasL (P =
.042), negative bcl-2 (P = .0004), and PFNB (P <
.0001) were associated with a significantly worse
prognosis in overall survival. In clinical stages II
through IV, positive FasL (P = .04), negative bcl-2
(P = .002), and PFNB (P < .0001) had significant
associations with shorter disease-free survival and
positive FasL (P = .049), negative bcl-2 (P < .0001),
and PFNB (P < .0001) with worse overall survival.
© American Society for Clinical Pathology
Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021
Carcinoma of the uterine cervix is still a major problem
among gynecologic malignant neoplasms, despite the success
of cytologic techniques in diagnosing precursor lesions and
the early diagnosis of invasive disease.1 In Japan, the number
of patients with newly diagnosed cervical cancer has been
increasing again, after a decline, mostly among young age
groups.2 Although radiation therapy and surgery are the prin-
cipal therapeutic modalities for disease limited to the pelvis,
patients with advanced disease or recurrence after locoregion-
al therapy depend on systemic treatment for any hope of sur-
vival. Because chemotherapy is not considered as effective as
radiation therapy and surgery, information about factors con-
tributing to a worse prognosis for patients should be valuable
for creating new therapeutic strategies.3
Fas ligand (FasL; CD95L), one of the tumor necrosis fac-
tor family proteins, is a type II membrane protein. It is known
that FasL expressed on activated T lymphocytes or natural
killer cells triggers apoptosis on the cells that express Fas
(CD95), one of the tumor necrosis factor receptor families.4
FasL-expressing tumors have been reported to have a signifi-
cantly worse prognosis.5-12 It is hypothesized that these tumor
cells can escape from immune surveillance via a counterattack
on activated T cells and natural killer cells that express Fas.13
One of the oncoproteins known to inhibit apoptosis is bcl-
2. This is thought to regulate the permeability transition pore
in the mitochondrial membrane in cooperation with Bax,
which also is one of the bcl-2 family proteins.14-16 Although it
is theoretically paradoxical, many studies have reported the
usefulness of bcl-2 as a favorable prognostic indicator.17-22
In the present study, protein expression of FasL and bcl-
2 was examined in cervical carcinoma. Prognostic signifi-
cance of the findings was evaluated.
Am J Clin Pathol 2005;123:879-885 879
879 DOI: 10.1309/0773N4Q3GFP34J5V 879
Munakata et al / FASL AND BCL-2 IN CERVICAL CARCINOMA

Materials and Methods
Cases
We selected 125 cases, diagnosed as invasive uterine cervi-
cal carcinoma from January 1994 to July 2004, from the file of
the Department of Pathology, Kansai Rosai Hospital,
Amagasaki, Japan. The study protocol was approved by the
Kansai Rosai Hospital Human Ethics Review Committee, and a
signed consent form was obtained from each subject for the use
of tissue samples in this study. A profile of the cases is given in
❚Table 1❚. A profile of 101 cases, chosen from January 1994 to
Simple Stain MAX PO (MULTI) kit (Nichirei, Tokyo, Japan).
We used 3,3'-diaminobenzidine as the chromogen. A
tris(hydroxymethyl)aminomethane buffer solution was substi-
tuted for the primary antibody in the negative control sample.
An ovarian cancer specimen that was known to express FasL
was used for a positive control specimen. A tonsil sample was
used for the positive control specimen for bcl-2.
Staining intensity was scored as 0 (negative), 1 (1+), 2
(2+), or 3 (3+) and the percentage of positive cells as 0 (0%),
1 (<10%), 2 (10%-50%), or 3 (>50%). When the value of the
percentage positive score times the intensity score was 2 or

Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021

December 2000 to assess prognostic significance, also is given. more, the case was regarded as positive. Lymphocyte infiltra-
Therapeutic modalities for the 101 cases are shown in ❚Table 2❚. tion in the tumor was graded as 0 (negative), 1 (1+), 2 (2+), or
Evaluation of lymph node metastatic status was done in 3 (3+) on H&E-stained sections.
116 available cases. Histologic examination was the method
of detection of lymph node metastasis in 61 cases, magnetic Statistical Analysis
resonance imaging in 27 cases, computed tomography in 18 Statistical analysis was done with Statcel software (OMS,
cases, ultrasonography in 4 cases, and palpation in 6 cases. Of Japan) and StatView 5.0 software (SAS Institute, Cary NC).
the cases in the 1994-2000 group, 85 were available for eval-
uation of tumor recurrence. Disease-free survival among the
group of cases in clinical stages I through IV was analyzed in ❚Table 2❚
97 available cases. In clinical stages II through IV, 68 and 71 Therapeutic Modalities for 101 Patients
cases were available for evaluation of disease-free survival
Operative Procedure Radiation Chemotherapy No. of Patients
and overall survival, respectively.
Conization No No 1
Methods Lymphadenectomy Yes Yes 1
Hysterectomy
We used 3-µm, formalin-fixed, paraffin-embedded tissue Simple total No No 1
Yes No 1
sections for immunohistochemical analysis. For primary anti- Yes Yes 1
bodies, we used FasL (NOK-2, dilution 1:50; Pharmingen, Semiradical No No 6
San Diego, CA) and bcl-2 (clone 124, dilution 1:100; DAKO, Yes No 1
Radical No No 3
Tokyo, Japan). Briefly, endogenous peroxidase was blocked No Yes 3
by 3% hydrogen peroxide following deparaffinization. For Yes No 10
Yes Yes 17
bcl-2, the antigen was retrieved by microwave oven heating None Yes No 36
(900 W, 15 minutes) in citrate buffer (pH 7.0, IATRON, Yes Yes 20
Total 101
Tokyo, Japan). Sections were processed using the Histofine

❚Table 1❚
Clinicopathologic Features for Patients*

Jan 1994-Dec 2000 (n = 101) Jan 1994-Jul 2004 (n = 125)

Mean ± SD age (range) (y) 61.2 ± 15.2 (23-95) 58.6 ± 15.7 (23-95)
Clinical stage
I 30 43
II 39 48
III 23 24
IV 9 10
Histologic diagnosis
Squamous cell carcinoma 89 107
Adenosquamous carcinoma 5 5
Adenocarcinoma 7 13
Follow-up period (mo)
Mean ± SD 40.7 ± 29.0 ND
Median 37 ND
Range 1-104 ND

ND, not done.

* Data are given as number of patients unless otherwise indicated.

880 Am J Clin Pathol 2005;123:879-885 © American Society for Clinical Pathology

880 DOI: 10.1309/0773N4Q3GFP34J5V
 Anatomic Pathology / ORIGINAL ARTICLE

The comparison of the distribution of expression of each fac- bcl-2 Staining

tor in the clinical stages was done by using the Mann-Whitney
U test. FasL and/or bcl-2 status and the histologic category
were compared by using the χ2 test for independence.
Associations between lymph node metastasis or recurrence
and the status of each factor were compared by using the
Fisher exact probability test. Actuarial survival curves of the
patients were constructed by using the Kaplan-Meier method,
and the significance of the differences was estimated by using
the log-rank test. The Cox regression model was used for uni-
variate and multivariate analyses.
bcl-2 staining was observed in the cytoplasm of carcinoma
cells and some infiltrating lymphocytes ❚Image 1B❚. Positive
staining was observed in 95 (76.0%) of 125 cases. Of 107 cases
of squamous cell carcinoma, 83 (77.6%) were positive, as were
6 (46%) of 13 cases of adenocarcinoma and 5 (100%) of 5 cases
of adenosquamous carcinoma. Expression of bcl-2 had no sig-
nificant association with histologic type (P = .13) or clinical
stage (P = .37). The lymphocyte infiltration score also had no
significant association with bcl-2 expression (P = .33).
Analysis of Lymph Node Metastasis and Expression of

Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021

Factors
Association of lymph node metastasis and expression of
Results
each factor was calculated ❚Table 3❚. Evaluation of lymph node

Fas Ligand
FasL expression was observed in cancer cells and in ❚Table 3❚
some lymphocytes with membranous and cytoplasmic Associations of Variables With Lymph Node Metastasis and
Recurrence*
staining ❚Image 1A❚. It showed a heterogeneous staining
pattern. Of 125 cases, 93 (74.4%) were regarded as positive. Variable
Of 107 cases of squamous cell carcinoma, 78 (72.9%) were Parameter Fas Ligand bcl-2 PFNB
positive, as were 12 (92%) of 13 cases with adenocarcino-
ma and 3 (60%) of 5 cases with adenosquamous carcinoma. Lymph node metastasis
(n = 116)
FasL was positive in 31 (72%) of 43, 35 (73%) of 48, 18 Positive 23/9 22/10 9/23
(75%) of 24, and 9 (90%) of 10 cases in clinical stages I, II, Negative 63/21 64/20 12/72
P .73 .41 .084
III, and IV, respectively. No associations were found Recurrence (n = 85)
between FasL expression and histologic type (P = .66) or Positive 19/4 15/8 7/16
Negative 44/18 52/10 6/56
clinical stage (P = .45). In addition, the lymphocyte infiltra- P .21 .061 .018
tion score and FasL expression did not show a significant
PFNB, positive Fas ligand and negative bcl-2.
association (P = .34). * Data are given as positive cases/negative cases.

A B

❚Image 1❚ Immunohistochemical analysis for Fas ligand and bcl-2 in cervical carcinoma. A, Fas ligand is expressed on the
membrane and in the cytoplasm of squamous cell carcinoma (immunoperoxidase, ×200). B, The cytoplasm of the tumor cells
and infiltrating lymphocytes show bcl-2 (immunoperoxidase, ×200).

© American Society for Clinical Pathology Am J Clin Pathol 2005;123:879-885 881

881 DOI: 10.1309/0773N4Q3GFP34J5V 881
Munakata et al / FASL AND BCL-2 IN CERVICAL CARCINOMA

metastasis was done in 116 available cases. No significant
associations were found between lymph node metastasis and
FasL or bcl-2 (Table 3). Tumors with positive FasL expression
and negative bcl-2 expression (PFNB) had a marginally signif-
icant association with lymph node metastasis compared with
tumors other than those designated PFNB (P = .084).
Analysis of Recurrence and Expression of Factors
When association of tumor recurrence with each factor
was evaluated, only the PFNB group showed a significant
association (Table 3). No significant associations were found
ratios of 6.76 in disease-free survival and 14.7 in overall sur-
vival. The hazard ratios for PFNB were the highest of all fac-
tors in disease-free and overall survival on univariate analysis.
Clinical stage no longer had a significant difference in disease-
free and overall survival when cases in clinical stage II were
compared with those in clinical stages III and IV combined.
Multivariate analysis of cases in clinical stages I through
IV revealed that negative bcl-2 was associated significantly with
a shorter disease-free survival (Table 4). Clinical stage, positive
FasL, and negative bcl-2 had a significant association with a
worse overall survival. In cases in clinical stages II through IV,

Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021

between recurrence and FasL or bcl-2 (Table 3).
Analysis of Survival and Expression of Factors
Univariate analysis of disease-free and overall survival is
given in ❚Table 4❚. When cases in clinical stages I through IV
were analyzed, negative bcl-2 and PFNB were associated sig-
nificantly with decreased disease-free survival. Cases with pos-
itive FasL, negative bcl-2, and PFNB had a significantly short-
er overall survival ❚Figure 1❚. When cases in combined clinical
stages I and II were compared with those in stages III and IV,
cases in advanced stages had significantly worse overall sur-
vival. Because it is especially important to know the factors
associated with a poor prognosis in advanced stages, we con-
ducted an analysis of survival among cases in clinical stages II
through IV ❚Figure 2❚. As shown in Table 4, positive FasL, neg-
ative bcl-2, and PFNB were associated significantly with short-
er disease-free and overall survival. PFNB showed hazard
positive FasL and negative bcl-2 had significantly shorter dis-
ease-free and overall survival on multivariate analysis.
Discussion
Clinical stage is a major prognostic factor in cervical car-
cinoma.23 It would be intriguing to know factors associated
with shorter survival in cases of advanced stage disease, which
frequently cannot be controlled by locoregional therapy and,
therefore, has a poor prognosis.
Many malignant tumors have been reported to express
FasL. Hahne et al13 reported that melanoma cells expressing
FasL grew more rapidly in wild-type mice than in Fas-
mutant lpr mice. From their experiment, it was expected that
FasL expression in tumor cells would affect tumor progres-
sion and the prognosis of patients. In fact, FasL expression

❚Table 4❚
Univariate and Multivariate Analyses of Survival for Each Variable

Disease-Free Survival Overall Survival

Variable No. of Cases P Hazard Ratio No. of Cases P Hazard Ratio

Univariate analysis
Stage I-IV 97 101
Stage* .09 .019 2.97
Fas ligand–positive .16 .042 6.21
bcl-2–negative .035 2.25 .0004 4.70
PFNB .0025 3.21 <.0001 7.52
Stage II-IV 68 71
Stage† .73 .34
Fas ligand–positive .040 1.02 .049 5.92
bcl-2–negative .002 3.34 <.0001 7.46
PFNB <.0001 6.76 <.0001 14.7
Multivariate analysis
Stage I-IV 97 101
Stage* .051 .0084 5.86
Age .53 .21
Fas ligand–positive .095 .032 15.9
bcl-2–negative .01 4.88 .0003 22.2
Stage II-IV 68 71
Stage† .55 .38
Age .91 .14
Fas ligand–positive .0094 7.92 .0047 18.1
bcl-2–negative .0012 7.57 <.0001 25.0

PFNB, positive Fas ligand and negative bcl-2.

* Stages I and II vs III and IV.
† Stage II vs stages III and IV.

882 Am J Clin Pathol 2005;123:879-885 © American Society for Clinical Pathology

882 DOI: 10.1309/0773N4Q3GFP34J5V
 Anatomic Pathology / ORIGINAL ARTICLE

A B

1.00 1.00
Cumulative Overall Survival

Cumulative Overall Survival

FasL (–) (n = 24)
0.80 0.80
bcl-2 (+) (n = 80)
FasL (+) (n = 77)
0.60 0.60

0.40 0.40 bcl-2 (–) (n = 21)

0.20 0.20

Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021

0.00 0.00

0 20 40 60 80 100 120 0 20 40 60 80 100 120

Follow-up (mo) Follow-up (mo)

C ❚Figure 1❚ Overall survival curves for 101 patients with

cervical carcinoma, clinical stages I through IV, in relation to
1.00 Fas ligand (FasL) (A), bcl-2 (B), and combined FasL and bcl-2
Cumulative Overall Survival

(C). Survival was analyzed using the Kaplan-Meier method,

0.80 FasL (+), bcl-2 (+) and comparison of study groups was performed using the
FasL (–), bcl-2 (–) log-rank test (A, P = .042; B, P = .0004; C, P < .0001).
0.60 FasL (–), bcl-2 (+) (n = 85)

0.40
FasL (+), bcl-2 (–) (n = 16)
0.20

0.00

0 20 40 60 80 100 120
Follow-up (mo)

had a significant association with lymph node metastasis in
breast cancer, gastric cancer, and colorectal cancer.5,6,9
Furthermore, FasL expression was associated significantly
with a poor prognosis in ovarian cancer.7 Cervical adenocar-
cinoma had a similar trend, although the number of patients
was small.12 Reimer et al10 reported that a FasL/Fas ratio
greater than 1 had prognostic significance in disease-free
and overall survival in tamoxifen-resistant, hormone recep-
tor–positive breast cancer. Moreover, Sjöström et al11 report-
ed a significant association of FasL and bcl-2 with overall
survival in univariate and multivariate analyses in breast can-
cer. In our study, univariate and multivariate analyses
showed that immunohistochemical expression of FasL had a
significant association with a poor prognosis in overall sur-
vival for patients grouped in clinical stages I through IV and
II through IV and in disease-free survival for those grouped
in stages II through IV.
It also has been reported that bcl-2 has prognostic signifi-
cance in many malignant tumors, including cervical cancer.17-22
Every report showed that negative bcl-2 expression con-
tributed to shorter survival. In the present study, univariate and
multivariate analyses revealed that patients with negative bcl-
2 expression had decreased disease-free and overall survival
among groups of clinical stages I through IV and II through
IV. Because the bcl-2 protein is known to have an antiapoptot-
ic effect and confer a survival advantage in preclinical mod-
els,14-16 this phenomenon is paradoxical. Two possible expla-
nations have been reported in breast cancer. One is that bcl-2
expression is correlated with estrogen receptor expression.24
Another explanation is that negative bcl-2 expression is asso-
ciated with an increased proliferation rate.24 Although nega-
tive estrogen receptor expression and a high proliferation rate
are known to be correlated with a worse prognosis, the exact
mechanisms of these correlations are unknown.
Multivariate analysis of cases in clinical stages I through
IV revealed that bcl-2 had prognostic significance in disease-
free survival. Clinical stage and FasL also were significant
prognostic factors along with bcl-2 in overall survival. In cases
in clinical stages II through IV, FasL and bcl-2 had significant
associations with disease-free and overall survival. FasL and
bcl-2 had meaningful associations in survival in advanced cer-
vical carcinoma even in multivariate analysis.

© American Society for Clinical Pathology Am J Clin Pathol 2005;123:879-885 883

883 DOI: 10.1309/0773N4Q3GFP34J5V 883
 Munakata et al / FASL AND BCL-2 IN CERVICAL CARCINOMA

A B

1.00 1.00
bcl-2 (+) (n = 59)
Cumulative Overall Survival

Cumulative Overall Survival

0.80 0.80
FasL (–) (n = 16)

0.60 0.60
FasL (+) (n = 55)
0.40 0.40
bcl-2 (–) (n = 12)
0.20 0.20

Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021

0.00 0.00
0 20 40 60 80 100 120 0 20 40 60 80 100 120
Follow-up (mo) Follow-up (mo)

❚Figure 2❚ Overall survival curves for 71 patients with cervical

C
carcinoma, clinical stages II through IV, in relation to Fas
1.00 ligand (FasL) (A), bcl-2 (B), and combined FasL and bcl-2 (C).
Cumulative Overall Survival

Survival was analyzed using the Kaplan-Meier method, and

0.80 comparison of study groups was performed using the log-
FasL (+), bcl-2 (+)
FasL (–), bcl-2 (–) rank test (A, P = .049; B, P < .0001; C, P < .0001).
0.60
FasL (–), bcl-2 (+) (n = 62)
0.40

0.20 FasL (+), bcl-2 (–) (n = 9)

0.00

0 20 40 60 80 100 120
Follow-up (mo)

Sjöström et al11 showed that breast cancer with high bcl-
2 expression and low FasL expression was associated with
the best overall survival. In the present study, we observed a
similar phenomenon. Patients with PFNB had the worst dis-
ease-free and overall survival among patients grouped by
clinical stages I through IV and II through IV. Although the
number of cases is small, no patient in clinical stages II
through IV with PFNB survived beyond 40 months. It is sug-
gested that PFNB would be a powerful prognostic indicator
in advanced cervical cancer.
In the present study, PFNB had a marginally significant
association with lymph node metastasis (P = .084), although
only 52.6% of the cases were proven histologically.
Practically, it often is difficult to prove lymph node metastasis
histologically in patients with advanced disease because those
patients usually are treated by radiation or chemotherapy
rather than surgery. Further study is required to know the true
association of these factors with lymph node metastasis.
Positive FasL expression and negative bcl-2 expression
had prognostic significance in invasive cervical carcinoma in
univariate and multivariate analyses. Furthermore, PFNB was
thought to be a powerful prognostic indicator in advanced
cervical carcinoma. These factors would be useful to select
cases with a poor prognosis in advanced cervical carcinoma.
Further study is required to create new therapeutic modalities
that would be beneficial to those cases.
From the Departments of 1Pathology, and 2Obstetrics and
Gynecology, Kansai Rosai Hospital, Amagasaki, Japan.
Supported in part by a grant aid from the Labour Welfare
Corporation (2000), Tokyo, Japan.
Address reprint requests to Dr Munakata: Dept of Pathology,
Kansai Rosai Hospital, 3-1-69, Inabasou, Amagasaki, Hyogo 660-
8511, Japan.
Acknowledgments: We thank Ryuichi Yoshino, Ayami
Horimoto, Kazuko Oku, Michiaki Yamane, and Kenji Sugio, for
technical assistance.
References
1. Wright TC, Ferenczy A, Kurman RJ. Carcinoma and other
tumors of the cervix. In: Kurman RJ, ed. Blaustein’s Pathology
of the Female Genital Tract. 4th ed. New York, NY: Springer-
Verlag; 1994:279-326.
2. Ito T, Ishizuka T, Suzuki K, et al. Cervical cancer in young
Japanese women. Arch Gynecol Obstet. 2000;264:68-70.

884 Am J Clin Pathol 2005;123:879-885 © American Society for Clinical Pathology

884 DOI: 10.1309/0773N4Q3GFP34J5V

3. Farias-Eisner R, Shapter AP, Casey AC, et al. Cervix, vagina,
and vulva. In: Abeloff MD, Armitage JO, Lichter AS, et al,
eds. Clinical Oncology. New York, NY: Churchill Livingstone;
1995:1527-1567.
4. Nagata S, Golstein P. The Fas death factor. Science.
1995;267:1449-1456.
5. Mann B, Gratchev A, Bohm C, et al. FasL is more frequently
expressed in liver metastases of colorectal cancer than in
matched primary carcinomas. Br J Cancer. 1999;79:1262-1269.
6. Mottolese M, Buglioni S, Bracalenti C, et al. Prognostic
relevance of altered Fas (CD95)-system in human breast
cancer. Int J Cancer. 2000;89:127-132.
7. Munakata S, Enomoto T, Tsujimoto M, et al. Expressions of
Fas ligand and other apoptosis-related genes and their
prognostic significance in epithelial ovarian neoplasms. Br J
Cancer. 2000;82:1446-1452.
8. Younes M, Schwartz MR, Ertan A, et al. Fas ligand expression
in esophageal carcinomas and their lymph node metastases.
Cancer. 2000;88:524-528.
9. Nagashima H, Mori M, Sadanaga N, et al. Expression of Fas
ligand in gastric carcinoma relates to lymph node metastasis.
Int J Oncol. 2001;18:1157-1162.
10. Reimer T, Koczan D, Müller H, et al. Tumour Fas ligand:Fas
ratio greater than 1 is an independent marker of relative
resistance to tamoxifen therapy in hormone receptor positive
breast cancer. Breast Cancer Res. 2002;4:R9.
11. Sjöström J, Blomqvist C, Boguslawski K, et al. The predictive
value of bcl-2, bcl-xL, bag-1, Fas, and FasL for chemotherapy
response in advanced breast cancer. Clin Cancer Res.
2002;8:811-816.
12. Kase H, Aoki Y, Tanaka K. Fas ligand expression in cervical
adenocarcinoma: relevance to lymph node metastasis and
tumor progression. Gynecol Oncol. 2003;90:70-74.
13. Hahne M, Rimoldi D, Schröter M, et al. Melanoma cell
expression of Fas (Apo-1/CD95) ligand: implications for
tumor immune escape. Science. 1996;274:1363-1366.
© American Society for Clinical Pathology
Anatomic Pathology / ORIGINAL ARTICLE
14. Hockenbery D, Nuñez G, Milliman C, et al. bcl-2 is an inner
mitochondrial membrane protein that blocks programmed cell
death. Nature. 1990;348:334-336.
15. Oltvai ZN, Milliman CL, Korsmeyer SJ. bcl-2 heterodimerizes
in vivo with a conserved homolog, Bax, that accelerates
programmed cell death. Cell. 1993;74:609-619.
16. Kroemer G. The proto-oncogene bcl-2 and its role in
regulating apoptosis. Nat Med. 1997;3:614-620.
17. Pezzella F, Turley H, Kuzu I, et al. bcl-2 protein in non–small-
cell lung carcinoma. N Engl J Med. 1993;329:690-694.
18. Joensuu H, Pylkkänen L, Toikkanen S. bcl-2 protein
expression and long-term survival in breast cancer. Am J
Pathol. 1994;145:1191-1198.
Downloaded from https://academic.oup.com/ajcp/article/123/6/879/1759572 by guest on 09 December 2021
19. Diebold J, Baretton G, Felchner M, et al. bcl-2 expression, p53
accumulation, and apoptosis in ovarian carcinomas. Am J Clin
Pathol. 1996;105:341-349.
20. Crawford RA, Caldwell C, Iles RK, et al. Prognostic
significance of the bcl-2 apoptotic family of proteins in
primary and recurrent cervical cancer. Br J Cancer.
1998;78:210-214.
21. Itoi T, Yamana K, Bilim V, et al. Impact of frequent bcl-2
expression on better prognosis in renal cell carcinoma patients.
Br J Cancer. 2004;90:200-205.
22. Saito T, Takehara M, Tanaka R, et al. Correlation between
responsiveness of neoadjuvant chemotherapy and apoptosis-
associated proteins for cervical adenocarcinoma. Gynecol
Oncol. 2004;92:284-292.
23. Follen M, Levenback CF, Iyer RB, et al. Imaging in cervical
cancer. Cancer. 2003;98(9 suppl):2028-2038.
24. Slooten HJ, Vijver MJ, Velde CJH, et al. Loss of bcl-2 in
invasive breast cancer is associated with high rates of cell
death, but also with increased proliferative activity. Br J
Cancer. 1998;77:789-796.
Am J Clin Pathol 2005;123:879-885 885
885 DOI: 10.1309/0773N4Q3GFP34J5V 885