Gynecologic Oncology 122 (2011) 238–241

Contents lists available at ScienceDirect

Gynecologic Oncology

j ourna l homepage: www.e lsev ie r.com/ locate /ygyno

The impact of pretreatment thrombocytosis and persistent thrombocytosis afteradjuvant chemotherapy in patients with advanced epithelial ovarian cancer

Maria Lee, Sang Wun Kim, Eun Ji Nam, Ga Won Yim, Sunghoon Kim, Young Tae Kim ⁎

Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Yonsei University College of Medicine, Seoul, Republic of Korea

⁎ Corresponding author at: Division of GynecologObstetrics and Gynecology, Yonsei University College134 Shinchon-dong, Seodaemun-gu, Seoul, 120-752, Rep8357.

E-mail address: ytkchoi@yuhs.ac (Y.T. Kim).

0090-8258/$ – see front matter © 2011 Elsevier Inc. Aldoi:10.1016/j.ygyno.2011.04.012

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 4 March 2011Accepted 12 April 2011Available online 4 May 2011

Keywords:ThrombocytosisPlatelet countEpithelial ovarian cancer

Objective. To evaluate the impact of both pretreatment thrombocytosis, and platelet count reduction post-adjuvant chemotherapy, on survival in patients with advanced epithelial ovarian cancer.

Methods. Records of 179 women who underwent cytoreductive surgery for FIGO stage III or IV epithelialovarian cancer and received six cycles of platinum/paclitaxel-based chemotherapy between July1998 andMarch 2009 were retrospectively reviewed. Platelet ratio was defined as the preoperative platelet countdivided by the platelet count after chemotherapy. The prognostic significance of thrombocytosis and plateletratio, together with various clinicopathological factors, were evaluated by multivariate analysis.

Results. Sixty-two of 179 (34.6%) patients had thrombocytosis at primary diagnosis. Patients with

preoperative thrombocytosis had greater elevations of CA-125 (pb0.0001) and a greater volume of ascites(p=0.007). On multivariate analysis, thrombocytosis and CA-125 elevation retained significance asindicators of poor prognosis in patients with stage III or IV disease. In patients with normal CA-125 afterchemotherapy, a high platelet ratio was an independent risk factor for reduced survival (p=0.05).

Conclusions. Preoperative thrombocytosis and a high platelet ratio appear to be poor prognostic factors ofsurvival in patients with advanced epithelial ovarian cancer whowere treated with cytoreductive surgery andadjuvant platinum/paclitaxel-based chemotherapy.

© 2011 Elsevier Inc. All rights reserved.

Introduction

Thrombocytosis has been demonstrated to be an unfavorableprognostic factor in a number of malignancies, including lung, gastric,breast, and pancreatic cancers [1–4]. In gynecologic cancers, throm-bocytosis appears to be of prognostic value in selected patients [5]. Inovarian cancer, elevated levels of interleukin-6 (IL-6) in ascites andcyst fluids have been associated with thrombocytosis [6,8]. IL-6 andgranulocyte-macrophage colony stimulating factor (GM-CSF) maystimulate thrombopoiesis and ultimately provoke thrombocytosis[6,7]. Administration of cytotoxic drugs results in a lowering of theplatelet count. Currently, it is unclear whether decreasing plateletcounts by administration of cytotoxic drugs, or inhibiting tumor-induced platelet activation using cyclooxygenase-2 inhibitors ornonsteroidal anti-inflammatory drugs, has a place in the treatmentof malignant disease [9]. Although this is an important clinical issue,there are no studies assessing changes in platelet count followingchemotherapy, and the clinical impact of platelet reduction on ovariancancer progression is unknown.

ic Oncology, Department ofof Medicine, 250 Seongsanno,ublic of Korea. Fax: +82 2 313

l rights reserved.

In this study, we investigated the correlation between preopera-tive thrombocytosis and clinical characteristics in advanced epithelialovarian cancer (EOC) and assessed the prognostic impact of changesin platelet count following adjuvant chemotherapy.

Materials and methods

Patients

All patient data for the current studywere retrieved from a databaseof 393 patientswhowere diagnosedwith epithelial ovarian cancer afterstaging laparotomy at Yonsei University Health System betweenJuly1998 and March 2009. In total, 179 patients were included in thecurrent study. This study was approved by the Institutional ReviewBoard of the Yonsei University College of Medicine.

Eligibility criteria were as follows: i) histological confirmation ofepithelial ovarian cancer; ii) International Federation of Gynecologyand Obstetrics (FIGO) stage III or IV disease; iii) completion of sixcycles of primary adjuvant chemotherapy following debulking sur-gery; and iv) Eastern Cooperative Oncology Group (ECOG) perfor-mance status of 0–2 at the time of diagnosis.

Chemotherapy was started within 3 weeks after surgery andconsisted of 6–9 cycles of paclitaxel (175 mg/m2) and carboplatin(area under curve [AUC]=5) every 3 weeks. Baseline patient charac-teristics included patient age, Gynecologic Oncology Group (GOG)

239M. Lee et al. / Gynecologic Oncology 122 (2011) 238–241

performance status, FIGO stage, ascites volume, tumor histology, andresidual tumor size after surgery.

Clinical and laboratory data collection

For all study subjects, platelet counts and CA-125 levels wererecorded at primary diagnosis up to 1 week prior to operation. Aftersix cycles of primary adjuvant chemotherapy, platelet counts andCA125 were re-measured. Platelet counts were analyzed using anADVIA 120/2120 Hematology system (Bayer HealthCare, DiagnosticsDivision, Tarrytown, NY, USA) and CA-125 levels were measuredusing a CA125 II ECLIA electrochemiluminescence immunoassay onthe Roche/Hitachi Modular Analytics E170 (Roche Diagnostics, Tokyo,Japan). Thrombocytosis was defined as a platelet count of at least40×104/mm3. The platelet ratio was defined as the preoperativeplatelet count divided by the platelet count after chemotherapy.

Overall survival (OS) was assessed from diagnosis to the date ofdeath from any cause, or to the date of last contact. Progression-freesurvival (PFS) wasmeasured from the completion of chemotherapy toeither the date of progression/relapse or to the date of last contact forpatients who were alive and demonstrated no disease progression.

Statistical analysis

Kaplan–Meier survival analysiswasused todetermine theunivariaterelationship of thrombocytosis and platelet ratio to overall andprogression-free survival. The log rank test was utilized to examinethe significance of the differences in survival between groups. Sub-sequently, multivariate analysis with Cox proportional hazards regres-sion was performed to determine which biomarkers predict PFS and OSafter adjusting for the effects of known prognostic factors (age, FIGOstage, and cell type). Generally, for all analyses a Pb0.05was consideredsignificant. All analyses were performed using SPSS version 18.0 (SPSSInc., Chicago, IL, USA).

Results

In total, 179 patientswere included in this study. Sixty-two (34.6%)patients had thrombocytosis at diagnosis. The characteristics of the

Table 1Patient characteristics.

Characteristics Non-thr(N=11

Age Mean (SDa) 54.3±1ECOGb score 0–1 89 (76

2 28 (23Stage IIIC 89 (84

IV 17 (16Ascites volume (mL) 600 (40Histology Serous 89 (76

Mucinous 13 (11Endometrioid 7 (6.0Clear 7 (6.0

Residual tumor after surgery b1 cm 34 (66≥1 cm 17 (33

Platelet, initial Median (range) 287 (11Platelet after CTxc Normal 88 (75

Thrombocytopenia 28 (23Thrombocytosis 1(0.9

Ratiod b2 87 (74≥2 30 (25

CA125, initial Median (range) 390 (3.1CA125 after CTx b35 86 (88

≥35 11 (11

a SD, standard deviation.b ECOG, Eastern Cooperative Oncology Group.c CTx, chemotherapy.d Ratio, pretreatment platelet/platelet after chemotherapy.

patients and their primary ovarian tumors are listed in Table 1. Thethrombocytosis and non-thrombocytosis groups were similar inpatient age, ECOG score, stage, histology, and residual tumor sizeafter surgery. However, patientswith thrombocytosis had significantlyhigher levels of preoperative CA-125 and ascites volume compared tothose without thrombocytosis.

Tumor stage, residual tumor size, and CA-125 levels after chemo-therapy were independent predictors of PFS (Table 2). On univariateanalysis, tumor histology, thrombocytosis, platelet ratio, platelet countsafter chemotherapy, and CA-125 levels after chemotherapy weresignificant prognostic factors of OS (Figs. 1 and 2). Thrombocytosis,persistent thrombocytosis after chemotherapy, and high CA-125 levelsafter chemotherapy, were independent poor prognostic factors of OS(Table 3). In the subset of patientswithCA-125b35after chemotherapy,those with a platelet ratio of 2 or greater showed a less-favorableprognosis than those with a platelet ratiob2 (48 vs. 66 months,p=0.05) (Fig. 3).

Discussion

In this study, 62 of 179 patients demonstrated thrombocytosisprior to surgery, and those with thrombocytosis had higher CA-125levels and ascites volume. These findings indicate that preoperativethrombocytosis is associated with more aggressive ovarian cancer.

The reported prevalence of thrombocytosis in patients withovarian cancer ranges from 22.4% to 62.5% [6,10,11]. Zeimet et al.[12] and Li et al. [11] found significantly higher rates of thrombocytosisin patients with stage III or IV than in those with stage I or II disease.However, others did not find statistically significant differences inthrombocytosis between ovarian cancer patients with stage III or IVdisease and those with early disease [9,13]. Zeimet et al. [12] did notfind significant differences in survival in stages I–IV disease with orwithout thrombocytosis. However, Li et al.[11] analyzed stages III andIV disease separately and found that disease-free and overall survivalrates were significantly affected by the presence of thrombocytosis inadvanced stage disease. The main difference between the two studieswas the proportion of patientwith stage III or IV disease included in thestudy groups (63% vs. 79.7%). In this study, 34.6% of patientswith stageIII or IV EOC had thrombocytosis before surgery. Stage, residual tumor

ombocytosis group7)

Thrombocytosis group(N=62)

p

1.1 55.3±11.2 0.716.1%) 48 (77.4%) 0.896.9%) 14 (22.6%).0%) 48 (82.8%) 0.829.0%) 10 (17.2%)–6000) 1750 (50–6000) 0.007.7%) 47 (77.0%) 0.422.2%) 3 (4.9%)%) 6 (9.8%)%) 5 (8.2%).7%) 21 (65.6%) 1.000.3%) 11 (34.4%)8–398) 512 (401–1054) b0.0001.2%) 49 (79.0%) 0.042.9%) 9 (14.5%)%) 4 (6.5%).4%) 28 (45.2%) b0.0001.6%) 34 (54.8%)-4703) 970 (6.2-11229) b0.0001.7%) 44 (81.5%) 0.230.3%) 10 (18.5%)

Table 2Univariate and multivariate analyses for determinants of progression-free survival.

Univariate analysis Multivariate analysis

p p HR (95% CI)

Ageb60 0.488FIGO stage III 0.001 0.002 0.192 (0.069–0.532)Pathology b0.0001 0.397 2.591 (0.287–23.418)Serous+endometrioidPlatelet count after CTx 0.346Platelet count,initialb40×104/mm3

0.324

Ratiob2 0.737Residual tumor sizeb1 cm 0.003 0.049 0.467 (0.218–0.998)CA-125 after CTxb35 μg/mL b0.0001 0.003 0.239 (0.094–0.611)AscitesN1000 mL 0.317

HR, hazard ratio; CI, confidence interval; LN, lymph nodes.

Fig. 2. Kaplan–Meier curves for overall survival of platelet ratio b2 group and plateletratio≥2 group. Median survival was 97 and 65 months, respectively (p=0.011).

240 M. Lee et al. / Gynecologic Oncology 122 (2011) 238–241

size after surgery, and tumor histology were not significantly differentbetween the thrombocytosis and non-thrombocytosis groups.

Aggressive ovarian cancer may be linked to the presence ofthrombocytosis. Our results showed that patients with thrombocy-tosis survived about 62 months, while those without the disorderlived for approximately 99 months, nearly 2 years longer.

Whether thrombocytosis is a reaction to more aggressive ovariancancer or the disorder itself exacerbates the disease is not clear.However, there is evidence suggesting that platelets play a role in thedevelopment of tumor metastases. Tumor cell–platelet interactionsmay influence the process of metastasis at different levels [14]. Tumorcell–platelet aggregates have been demonstrated during the initialarrest of tumor cells in the capillary vascular bed andplay an importantrole in hematogenous tumor spread [15]. Tumor cells can also directlyactivate platelets [16]. Platelets may protect tumor cells by coatingthem and blocking their antigenic determinants from the host'shumoral and cellular defensemechanisms [17,18]. Several preliminaryexperimental and clinical studies suggested an antitumor and anti-metastatic effect of anticoagulants [19]. Preclinical analysis of heparinprovides evidence for its anti-metastatic activity in a variety of animalmodels [20,21]. Anticoagulant therapyusing heparinmay affect cancer

Fig. 1. Kaplan–Meier curves for overall survival of pretreatment thrombocytosis groupand non-thrombocytosis group. Median survival was 99 and 62 months, respectively(p=0.005).

progression by inhibiting heparanase, blocking P- and L-selectin-mediated cell adhesion, and inhibiting angiogenesis [22].

Interestingly, the group with a platelet ratio 2 or greater showedpoor prognosis compared to the groupwith a ratio less than 2. Inducingthrombocytopenia would increase tumor vascular leakiness andfacilitate the effective delivery of chemotherapeutic agents to tumors.A recent study confirmed this hypothesis in tumor-bearingmice [23]. Inour results, thrombocytopenia after 6 cycles of chemotherapy was apoor prognostic factor compared to normal platelet count aftertreatment. It is difficult to accurately describe a reason for this finding.Of note, patients with thrombocytosis after chemotherapy had worseprognosis regardless of pretreatment platelet count. Thrombocytopeniamay have resulted in patients not receiving planned chemotherapy onschedule or at the full dose. In subgroup analysis of patientswith normalCA-125 after chemotherapy, a high platelet ratio was an independentpoor prognostic factor. Therefore, these results suggest that plateletcount may be useful as a tumor marker in ovarian cancer.

There are some limitations to our study. A relatively small numberof patientswere included in the analysis. In addition, this retrospectivechart reviewmay have unmeasured confounders. However, this is the

Table 3Univariate and multivariate analyses for determinants of overall survival.

Univariate analysis Multivariate analysis

p p HR (95% CI)

Ageb60 0.208FIGO stage III 0.148Pathology 0.024 0.125 0.397 (0.122–1.293)Serous+endometrioidPlatelet count,initialb40×104/mm3

0.005 0.027 0.297 (0.101–0.869)

Ratiob2 0.011 0.423 0.641 (0.216–1.904)Platelet count after CTx b0.0001

Normal 1.0 (reference)Thrombocytopenia 0.126 2.591 (0.765–8.773)Thrombocytosis 0.009 7.059 (1.632–30.537)

Residual tumor sizeb1 cm 0.388CA-125 after CTxb35 μg/mL 0.003 0.042 0.356 (0.131–0.964)AscitesN1000 mL 0.506

HR, hazard ratio; CI, confidence interval; LN, lymph nodes.

Fig. 3. Kaplan–Meier curves for overall survival of platelet ratio b2 group and plateletratio≥2 group in patients with CA-125b35 after chemotherapy. Median survival was66 and 48 months, respectively (p=0.050).

241M. Lee et al. / Gynecologic Oncology 122 (2011) 238–241

first report on change in platelet count in patients with advanced EOCwho received adjuvant chemotherapy and demonstrate its clinicalimpact.

In conclusion, preoperative thrombocytosis and a high plateletratio appear to be poor prognostic factors of survival in patients withadvanced EOC who were treated with cytoreductive surgery andadjuvant chemotherapy. Platelet count is a promising biomarker forEOC and further studies are needed to assess potential roles in EOC.

Conflict of interest statementThe authors declare that there are no conflicts of interest.

Acknowledgments

This study was supported by grants from the Yonsei UniversityResearch Fund of 2010 (6-2010-0006), a Faculty Research Grant ofYonsei University College of Medicine 2009 (6-2009-0127), and theNational Research Foundation of Korea Grant funded by the KoreanGovernment (7-2010-0264).

References

[1] Pedersen LM, Milman N. Diagnostic significance of platelet count and other bloodanalyses in patients with lung cancer. Oncol Rep 2003;10(1):213–6.

[2] Ikeda M, Furukawa H, Imamura H, Shimizu J, Ishida H, Masutani S, et al. Poorprognosis associated with thrombocytosis in patients with gastric cancer. AnnSurg Oncol 2002;9(3):287–91.

[3] Benoy I, Salgado R, Colpaert C, Weytjens R, Vermeulen PB, Dirix LY. Seruminterleukin 6, plasma VEGF, serum VEGF, and VEGF platelet load in breast cancerpatients. Clin Breast Cancer 2002;2(4):311–5.

[4] Schwarz RE, Keny H. Preoperative platelet count predicts survival after resectionof periampullary adenocarcinoma. Hepatogastroenterology 2001;48(41):1493–8.

[5] Gucer F, Tamussino K, Keil F, Balkanli-Kaplan P, Yuce MA. Thrombocytosis ingynecologic malignancies. Anticancer Res 2004;24(3b):2053–9.

[6] Gastl G, Plante M, Finstad CL, Wong GY, Federici MG, Bander NH, et al. High IL-6levels in ascitic fluid correlate with reactive thrombocytosis in patients withepithelial ovarian cancer. Br J Haematol 1993;83(3):433–41.

[7] Imai T, Koike K, Kubo T, Kikuchi T, Amano Y, Takagi M, et al. Interleukin-6 supportshuman megakaryocytic proliferation and differentiation in vitro. Blood 1991;78(8):1969–74.

[8] van der Zee AG, de Cuyper EM, Limburg PC, de Bruijn HW, Hollema H, Bijzet J, et al.Higher levels of interleukin-6 in cystic fluids from patients with malignant versusbenign ovarian tumors correlate with decreased hemoglobin levels and increasedplatelet counts. Cancer 1995;75(4):1004–9.

[9] Hernandez E, Donohue KA, Anderson LL, Heller PB, Stehman FB. The significance ofthrombocytosis in patients with locally advanced cervical carcinoma: a GynecologicOncology Group study. Gynecol Oncol 2000;78(2):137–42.

[10] Menczer J, Schejter E, Geva D, Ginath S, Zakut H. Ovarian carcinoma associatedthrombocytosis. Correlation with prognostic factors and with survival. Eur JGynaecol Oncol 1998;19(1):82–4.

[11] Li AJ, Madden AC, Cass I, Leuchter RS, Lagasse LD, Karlan BY. The prognosticsignificance of thrombocytosis in epithelial ovarian carcinoma. Gynecol Oncol2004;92(1):211–4.

[12] Zeimet AG, Marth C, Muller-Holzner E, Daxenbichler G, Dapunt O. Significance ofthrombocytosis in patients with epithelial ovarian cancer. Am J Obstet Gynecol1994;170(2):549–54.

[13] den Ouden M, Ubachs JM, Stoot JE, van Wersch JW. Whole blood cell counts andleucocyte differentials in patients with benign or malignant ovarian tumours. Eur JObstet Gynecol Reprod Biol 1997;72(1):73–7.

[14] Honn KV, Tang DG, Crissman JD. Platelets and cancer metastasis: a causalrelationship? Cancer Metastasis Rev 1992;11(3–4):325–51.

[15] Qian X, Tuszynski GP. Expression of thrombospondin-1 in cancer: a role in tumorprogression. Proc Soc Exp Biol Med 1996;212(3):199–207.

[16] Tuszynski GP, Gasic TB, Rothman VL, Knudsen KA, Gasic GJ. Thrombospondin, apotentiator of tumor cell metastasis. Cancer Res 1987;47(15):4130–3.

[17] Karpatkin S, Pearlstein E. Role of platelets in tumor cell metastases. Ann InternMed 1981;95(5):636–41.

[18] Pearlstein E, Ambrogio C, Karpatkin S. Effect of antiplatelet antibody on thedevelopment of pulmonary metastases following injection of CT26 colonadenocarcinoma, Lewis lung carcinoma, and B16 amelanotic melanoma tumorcells into mice. Cancer Res 1984;44(9):3884–7.

[19] HejnaM, Raderer M, Zielinski CC. Inhibition of metastases by anticoagulants. J NatlCancer Inst 1999;91(1):22–36.

[20] Niers TM, Klerk CP, DiNisio M, Van Noorden CJ, Buller HR, Reitsma PH, et al.Mechanisms of heparin induced anti-cancer activity in experimental cancermodels. Crit Rev Oncol Hematol 2007;61(3):195–207.

[21] SmorenburgSM,VanNoordenCJ. Thecomplexeffects ofheparinsoncancerprogressionand metastasis in experimental studies. Pharmacol Rev 2001;53(1):93–105.

[22] Borsig L. Antimetastatic activities of heparins and modified heparins. Experimentalevidence. Thromb Res 2010;125(Suppl 2):S66–71.

[23] Demers M, Ho-Tin-Noe B, Schatzberg D, Yang JJ, Wagner DD. Increased efficacy ofbreast cancer chemotherapy in thrombocytopenic mice. Cancer Res 2011;71(5):1540–9.