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.