International Journal of Gynecology & Obstetrics 75 Ž2001. 171᎐176
Article
Thymidine labeling index in epithelial ovarian cancer
Y. Salihoglua , A. Bilir b, A. Aydiner c,U , M. Erkan b, S. Tuzlali d, Y. Eralp c
a
Department of Obstetrics and Gynecology, Istanbul Faculty of Medicine, Istanbul, Turkey
Department of Histology and Embryology, Istanbul Faculty of Medicine Istanbul, Turkey
c
Department of Medical Oncology, Institute of Oncology, Istanbul, Capa, Turkey 34390
d
Department of Pathology, Istanbul Faculty of Medicine, Istanbul, Turkey
b
Received 28 February 2001; received in revised form 1 June 2001; accepted 7 June 2001
Abstract
Objecti¨ e: The aim of this study is to determine the thymidine labeling index and its prognostic role in patients with
ovarian cancer. Methods: Tumor cell proliferation in 32 patients with primary ovarian cancer admitted to Istanbul
Medical Faculty, Department of Obstetrics and Gynecology, between 1993 and 1997 was investigated using the
w 3 Hxthymidine labeling index ŽTLI.. TLI results were compared with other clinical and histopathologic prognostic
parameters. Results: The mean and median TLI values of the patients were 9.3" 6.2% and 9.20% Žrange:
0.4᎐23.0%., respectively. Sixteen patients showed high proliferation rates Žmean TLI: 14.3%.. These patients had an
overall survival rate of 46.7% at 3 years. The mean TLI level and overall survival at 3 years in the low proliferation
rate group were 4.4 and 68.8%, respectively. Patients with a high TLI had a significantly shorter survival compared to
those with a low TLI Ž P- 0.01.. There was tendency towards a higher TLI with advanced stage Ž P) 0.05.. However,
there was no statistically significant correlation between TLI and other prognostic parameters. Conclusion: TLI may
have a predictive value in determining the outcome of patients with ovarian cancer. Further larger scale studies are
needed before definite conclusions can be made about its role as a prognostic factor in this disease. 䊚 2001
International Federation of Gynecology and Obstetrics. All rights reserved.
Keywords: Ovarian tumor; Prognosis; Thymidine labeling index
U
Corresponding author. Tel.: q90-212-531-3100; fax: q90-216-330-3314.
0020-7292r01r$20.00 䊚 2001 International Federation of Gynecology and Obstetrics. All rights reserved.
PII: S 0 0 2 0 - 7 2 9 2 Ž 0 1 . 0 0 4 5 5 - 6
172
Y. Salihoglu et al. r International Journal of Gynecology & Obstetrics 75 (2001) 171᎐176
1. Introduction
Ovarian cancer cells show an unusual growth
potential. Various studies revealed a strong correlation between clinical and histopathologic
parameters w1᎐6x. Stage and largest residual mass
following surgical debulking have been recognized as major prognostic factors by some w2,3,7,8x;
while others have emphasized that grade is the
most important factor indicating a poorer outcome due to more aggressive biologic behavior
w3,7,9᎐11x. Investigators have recently been evaluating the role of tumor cell kinetics and proliferation index as potential prognostic parameters
w2,12᎐15x. Flow-cytometric S-phase analysis, which
has been used to determine the proliferative ca-
pacity of tumor cells, has been criticized due to
the lack of consistent data that would clarify its
role as a promising prognostic tool w15,16x. The
thymidine labeling index ŽTLI. has emerged as
one of the most reliable methods in assessing the
proliferation rate of certain cancer types w13,14x.
In this study, we investigated the correlation
between TLI and various prognostic parameters
in patients with ovarian cancer.
2. Patients and methods
2.1. Patient population
Tumor activity was evaluated in 32 patients
Table 1
Clinical, histopathological and cell kinetic values of the patients
Patient number
Age
Histology
Grade
Stage
Last status
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
35
40
47
46
41
44
53
52
50
51
55
53
50
50
51
50
56
63
57
67
67
62
60
59
62
59
74
66
64
64
65
67
Serous
Serous
Serous
Serous
Endometrioid
Serous
Endometrioid
Serous
Endometrioid
Serous
Endometrioid
Serous
Serous
Serous
Serous
Endometrioid
Serous
Mucinous
Mucinous
Serous
Serous
Mucinous
Mucinous
Serous
Serous
Serous
Serous
Serous
Serous
Serous
Endometrioid
Serous
2
2
2
3
2
2
3
3
2
1
1
2
1
2
2
1
1
1
1
3
3
1
3
2
3
2
2
3
2
3
2
2
III
III
IV
III
III
III
III
III
III
I
I
III
III
IV
IV
I
III
I
II
III
IV
III
IV
II
III
I
III
III
III
III
III
III
Alive
Dead
Dead
Dead
Alive
Alive
Alive
Dead
Dead
Alive
Alive
Alive
Alive
Dead
Dead
Alive
Dead
Alive
Alive
Dead
Dead
Dead
Alive
Alive
Dead
Alive
Alive
Alive
Alive
Dead
Alive
Dead
S-phase TLI %
20.21
11.92
14.81
8.6
9.4
11.38
1.17
4.64
15.84
0.4
10.61
10.98
12.91
11.19
12.33
1.1
6.06
2.59
8.16
2
23.04
21.55
1.86
12.88
9.01
8.52
6.57
1.69
1.8
18.73
5.66
9.96
Y. Salihoglu et al. r International Journal of Gynecology & Obstetrics 75 (2001) 171᎐176
with primary ovarian cancer admitted to the Istanbul University Istanbul Medical Faculty Department of Obstetrics and Gynecology, between
1993 and 1997. Patient characteristics are listed in
Table 1. Median age was 55 years, ranging
between 35 and 74 years. Six of the 32 patients
were premenopausal, nine patients were perimenopausal, and 17 patients were postmenopausal. Seven patients presented with stage I and
II disease, 20 patients had stage III, and five
patients had stage IV disease. Eight patients had
grade 1, 15 patients grade 2 and nine patients
grade 3 tumors. There were four mucinous, six
endometrioid and 22 serous papillary adenocarcinoma diagnosed. After a median follow-up period
of 36 months, 18 patients remained alive and 14
died due to tumor progression.
2.2. Determination of TLI
TLI was determined immediately after surgical
debulking on 36 tumor specimens obtained from
the resected ovaries. Following removal of fat and
the surrounding tissues, leaving a homogenous
tumoral mass, the tumor was minced into eight to
10 fragments of approximately 1 mm3 , and the
fragments were placed in 2 ml of 199 medium
with 20% fetal calf serum, streptomycin 100
mgrml, penicillin 100 Urml, 6 mCirml
w 3 Hxthymidine Žspecific activity 5 Cirmol.. They
were incubated for 1 h in agitation at 37⬚C in a
shaking water bath. After the incubating period,
the tumor fragments were briefly washed three
times in phosphate buffered solution and fixed in
buffered 10% formalin solution dehydrated in
alcohol, embedded in paraffin and 5-m sections
were cut. Slides were coated with emulsion film
ŽIlford K2. in a dark room and exposed at 4⬚C for
3᎐5 days. Autoradiographies were then developed
in D 19b for 5 min at 18⬚C and fixed in a standard
fixer. The slides were stained with hematoxylin
and eosin at 4⬚C.
The ratio of the labeled cells was determined
by counting a total of 1000᎐3000 cells for different specimens of the same tumor. When the
specimens were small enough to allow precursors
to penetrate, the counting was done throughout
the entire section. In the remaining cases, the
173
counting was limited to the periphery of the section. In all cases 20 grains overlying the nucleus
was necessary for a positive count. The TLI was
expressed as the percentage of epithelial cells
labeled with thymidine. Medium, serum and radiochemical material were supplied by Gibco
Laboratory ŽCat no: 1800-027., Biological Industries ŽIsrael Cat No: 04-121-IA. and Radiochemical Center ŽAmersham, United Kingdom TRA
120 specific activity 5 Cirmmol., respectively.
2.3. Statistical analysis
Statistical correlations between TLI and various prognostic factors were analyzed by the nonparametric analysis for several independent samples, the Kruskal᎐Wallis test. P-values less than
0.05 were considered significant. Patients were
placed in two groups based on their proliferative
index compared to the median TLI value for the
whole group, which was accepted as the cut-off
level. Patients with a lower ratio than the threshold TLI value were considered as the low proliferation rate group; while the remaining were
classified as the high proliferation rate group.
Comparison of the survival rate in these two
groups were performed by the log rank test.
3. Results
The TLI levels in 32 patients ranged between
0.4% and 23.04%. The mean and median TLI
values of the whole group were 9.3" 6.2% and
9.20% Žrange: 0.4᎐23.0%., respectively. The mean
and median TLI values of the 14 patients who
died during the 36 months follow-up period were
12.1 and 11.2%, respectively, while those of the
18 patients who were alive after the same period
were 7.1 and 7.4%, respectively. Data on TLI and
survival with respect to age and menopausal
groups are shown in Table 2. There was a tendency towards an inverse relation between age
and TLI; however, the difference was not significant. Furthermore, there was no significant correlation between TLI, age and survival. We did not
observe a significant relationship between the TLI
and various factors such as, grade, histology and
174
Y. Salihoglu et al. r International Journal of Gynecology & Obstetrics 75 (2001) 171᎐176
Table 2
Menopausal status, TLI values and overall survival at 3 years
Status
n
Age range
Mean TLI%
Ž"S.D..
Median TLI%
Žrange.
Survival rate
Ž%.
Premenopause
Perimenopause
Postmenopause
6
9
17
35᎐49
50᎐54
) 55
12.7 Ž4.3.
7.8 Ž5.9.
8.9 Ž6.8.
11.7 Ž8.6᎐20.2.
10.9 Ž0.4᎐15.8.
8.2 Ž1.7᎐23.0.
50
55.6
58.8
stage at presentation ŽTable 3.. Nevertheless,
patients with more advanced disease tended to
have higher TLI levels. When patients were classified into two groups with respect to TLI values,
16 patients were found to have proliferation rates
higher than the threshold value, which was designated as the median TLI of the whole group
Ž9.2%; range: 0.4᎐23.0%.. These patients had a
median TLI ratio of 12.6% Žrange: 9.4᎐23.0%.
and an overall survival rate of 46.7% at three
years. The median TLI level and overall survival
at 3 years in the low proliferation rate group were
3.6% Žrange: 0.4᎐9.0%. and 68.8%, respectively.
The patients with a lower TLI survived significantly longer than patients with a high TLI Ž P0.01. ŽTable 4.. A multivariate analysis could not
be performed due to the small sample size.
Table 3
Relationship between TLI values and histopathologic parameters
TLI
P
n Mean % Ž"S.D.. Median % Žrange.
Grade
1
2
3
8
15
9
7.9 Ž7.1.
10.9 Ž4.4.
7.9 Ž8.0.
7.1 Ž0.4᎐21.6.
11.2 Ž1.8᎐20.2.
4.6 Ž1.17᎐23.0.
0.19
Histology
Mucinous
4
Endometroid 6
Serous
22
8.5 Ž9.1.
7.3 Ž5.8.
9.9 Ž6.0.
5.4 Ž1.9᎐21.6.
7.5 Ž1.1᎐15.8.
10.5 Ž0.4᎐23.0.
0.51
6.3 Ž4.9.
9.5 Ž6.1.
12.7 Ž7.6.
8.2 Ž0.4᎐12.9.
9.2 Ž1.17᎐21.6.
12.3 Ž1.86᎐23.0.
0.17
Stage
I q II
III
IV
7
20
5
4. Discussion
In this study, we observed that a high proliferation rate of a tumor ensues shorter survival compared with tumors with a lower proliferation rate
Ž P- 0.05.. This observation may bring forward
certain clinical implications. In addition to the
generally accepted prognostic parameters like
stage, histological type, grade age and extent of
the residual tumor w2,17x, molecular and cellular
parameters are being evaluated more extensively
to determine their role in predicting the outcome
of patients with ovarian cancer. Although, these
prognostic indicators are not routinely used, they
may be of assistance in obtaining a thorough
understanding of the molecular biology of the
tumor per se leading to the development of better
therapeutic schemes and may also encourage further research in this field.
Table 4
The relationship between TLI values and 3-year survival
TLI value
n
Mean TLI %
Ž"S.D..
Median TLI %
Žrange.
Survival %
P
High group ŽTLI ) 9.2.
Low group ŽTLI F 9.2.
Total
16
16
32
14.2 Ž4.4.
4.4 Ž3.1.
9.3 Ž6.2.
12.6 Ž9.4᎐23.0.
3.6 Ž0.4᎐9.0.
9.2 Ž0.4᎐23.0.
46.7
68.8
56.3
- 0.01
Y. Salihoglu et al. r International Journal of Gynecology & Obstetrics 75 (2001) 171᎐176
Cell proliferation assays have been employed to
determine the prognosis in various other human
solid tumors w1,5x. In contrast to previous data
designating the importance of a high histological
grade as an indicator for clinical progression
w3,9᎐11x we did not observe a significant relationship between TLI and tumor grade. Compared
with other histologic types serous papillary tumors showed a relatively higher TLI rate, but the
difference was not significant. There was a tendency for higher TLI values with more advanced
stages, which is consistent with previous studies
reporting shorter survival in patients with extensive disease w3x. It is reasonable to argue that,
with a larger sample size, the association of TLI
and stage might have reached significance limits.
Furthermore, we observed a significant difference
in survival rates with respect to TLI levels, favoring those with a low index Ž P- 0.05.. Various
studies revealed that tumor cell proliferation rates
are very useful prognostic indicators for survival
w1,2,18᎐21x, especially in DNA diploid tumors
w22,23x. Although there are controversies regarding the relationship between survival and ploidy
w24,25x, it was previously demonstrated that
patients with higher S-phase fraction lived shorter
w24x. Other investigators claimed that higher Sphase fraction in ovarian tumors can predict early
recurrence and serve as an independent prognostic factor w28x, compared with cervical w26x and
endometrial w27x tumors. In addition to having a
potential predictive role in assessment of the outcome, tumor cell kinetics may enable us to determine the best treatment modality w14,29x. In
our study we determined that patients with a
lower TLI index survived significantly longer than
patients with a higher TLI index. Therefore, it
can be concluded that TLI index may serve as a
prognostic marker for epithelial ovarian cancer.
However, further large scale studies are required
before definite conclusions can be reached.
w3x
w4x
w5x
w6x
w7x
w8x
w9x
w10x
w11x
w12x
w13x
w14x
w15x
References
w1x Klemi PJ, Joensuu H, Maenpa AJ, Kiilholma P. Influence of cellular DNA content on survival in ovarian
carcinoma. Obstet Gynecol 1989;74:200᎐204.
w2x Silvestrini R, Daidone MG, Bolis G, Fontanelli R, Lan-
w16x
175
doni F, Androela S et al. Cell kinetics. A prognostic
marker in epithelial ovarian cancer. Gynecol Oncol
1989;35:15᎐18.
Redman JR, Petroni GR, Saigo PE, Geller NL, Hakes
TB. Prognostic factors in advanced ovarian carcinoma. J
Clin Oncol 1986;4:515᎐523.
Dembo AJ, Bush RS. Choise of postoperative therapy
based of prognostic factors. Int J Radiat Oncol Biol
Phys 1982;8:893᎐897.
Griffiths CT. Surgical resection of tumor bulk in the
primary treatment of ovarian carcinoma. Natl Cancer
Inst Monogr 1975;42:101᎐104.
Malkasian GD, Decker DG, Webb MJ. Histology of
epithelial tumors of the ovary: clinical usefulness and
prognostic significance of the histologic classification
and grading. Semin Oncol 1975;2:191᎐201.
Ozols RF, Garvin AJ, Costa J, Simon RM, Young RC.
Advanced ovarian cancer. correlation of histologic grade
with response to therapy and survival. Cancer
1980;45:572᎐581.
Bolis G, Marsoni S, Belloni C, Bianchi U, Bolis PF,
Bortolozzi G et al. Randomized comparison of cisplatin
with cyclophosphamidercisplatin and with cyclophosphamiderdoxorubicinrcisplatin in advanced ovarian
cancer. Lancet 1987;8555:353᎐355.
Richardson GS, Scully RE, Nikrui N, Nelson JH. Common epithelial cancer of the ovary. N Engl J Med
1985;312:415᎐424.
Rodenburg CJ, Corneliss CJ, Heintz PAM, Hermans J,
Fleuren GJ. Tumor ploidy as a major prognostic factor
in advanced ovarian cancer. Cancer 1987;59:317᎐323.
Friedlander ML, Hedley DW, Taylor IW, Russel P,
Coates AS, Tattersal MHN. Influence of cellular DNA
content on survival in advanced ovarian cancer. Cancer
Res 1984;44:397᎐400.
Iversen OE, Skaarland E. Ploidy assessment of benign
and malignant ovarian tumors by flow cytometry. Cancer
1987;60:82᎐87.
Karen J, Trope CG, Kristensen GB, Tveit TM, Peterson
EO. Evaluation of deoxyribonucleic acid ploidy and
s-phase fraction as prognostic parameters in advanced
epithelial ovarian carcinoma: a prospective study. Am J
Obstet Gynecol 1994;170:479᎐487.
Jelen I, Valente PT, Gautreaux L, Clark GM. Deoxyribonucleic acid and ploidy and s-phase fraction are not
significant prognostic factors for patients with cervical
cancer. Am J Obstet Gynecol 1994;171:1511᎐1518.
Gudmudsson TE, Hogberg T, Alm P, Anderson H,
Baldetorp B, Ferno M et al. The prognostic information
of DNA ploidy and s-phase fraction may vary with
histologic grade in endometrial carcinoma. Acta Oncol
1995;34:803᎐812.
Volm M, Bruggeman
A, Gunther
M, Kleine W, Pflei¨
¨
derer A, Vogt-Schaden M. Prognostic relevance of
ploidy, proliferation, and resistance-predictive tests in
ovarian carcinoma. Cancer Res 1985;45:5150᎐5185.
176
Y. Salihoglu et al. r International Journal of Gynecology & Obstetrics 75 (2001) 171᎐176
w17x Friberg LG, Noren H, Delle U. Prognostic value of
DNA ploidy and s-phase fraction in endometrial cancer
stage I and II: a prospective 5-year survival study. Gynecol Oncol 1994;53:64᎐69.
w18x Kallioniemi OP, Punnonen R, Mattila J, Lehtinen M,
Koivula T. Prognostic significance of DNA index, multiploidy, and s-phase fraction in ovarian cancer. Cancer
1988;61:334᎐339.
w19x Suzuki M, Ohwada M, Tamada T, Tsuru S. Thymidylate
synthase activity as a prognostic factor in ovarian cancer. Oncology 1994;51:334᎐338.
w20x Henriksen R, Strang P, Backstrom T, Wilander E,
Tribukait B, Oberk K. Ki-67 immunostaining and DNA
flow cytometry as prognostic factors in epithelial ovarian
cancers. Anticancer Res 1994;14:603᎐608.
w21x Gargano G, Catino A, Correale M, Lorusso V, Abbate
I, Izzi G et al. Prognostic factors in epithelial ovarian
cancer. Eur J Gynecol Oncol 1992;13ŽSuppl 1.:44᎐45.
w22x Kaleli S, Kosebay D, Bese T, DemirkŽran F, Oz UA,
Arvas M. Strong prognostic variable in endometrial carcinoma: flow cytometric s-phase fraction. Cancer
1997;79:44᎐51.
w23x Yamanaka S, Nagai N, Ohama K. Study of proliferating
cellular antigen ŽPCNA., nucleolar organizer regions
ŽNORs. and mitotic activity in epithelial ovarian tumors.
w24x
w25x
w26x
w27x
w28x
w29x
Nippon Sanka Fujinka Gakkai Zasshi ŽJapan .
1993;45:234᎐240.
Sahni K, Tribukait B, Einhorn N. Flow cytometric measurement of ploidy and proliferation in effusions of
ovarian carcinoma and their possible prognostic significance. Gynecol Oncol 1989;35:240᎐245.
Silvestrini R. Cell kinetics and therapeutic implications
in human tumors. Cell Prolif 1993;26:361᎐397.
Ozols RF, Young RC. Chemotherapy of ovarian cancer.
Semin Oncol 1984;11:251᎐263.
Daidone MG, Benini E, Valentinis B, Tomasic G, Bolis
G, Villa A et al. p53 expression, DNA content and cell
proliferation in primary and synchronous metastatic lesions from ovarian surface epithelial-stromal tumors.
Eur J Cancer 1996;32:1388᎐1393.
Evans MP, Webb MJ, Gaffey TA, Katzman JA, Suman
VJ, Hu TC. DNA ploidy of ovarian granulosa cell tumors. Lack of correlation between DNA index or proliferative index and outcome in 40 patients. Cancer
1995;75:2295᎐2298.
Reles AE, Gee C, Schellschmidt I, Schmider A, Unger
M, Friedmann W et al. Prognostic significance of DNA
content and s-phase fraction in epithelial ovarian carcinomas analyzed by image cytometry. Gynecol Oncol
1998;71:3᎐13.