Secondary Thrombocytosis as a poor prognostic indicator in ovarian carcinoma

Introduction : Secondary thrombocytosis has been identified in many solid tumors including ovarian carcinoma and has a poor prognostic value in many cases. Platelets and Platelet-related factors may contribute to metastasis, invasion and primary tumor growth. Objective : The objective of the present study is to determine the incidence of thrombocytosis in ovarian carcinoma and its importance as a poor prognosticator. Materials and methods : One sixty cases of epithelial ovarian tumors were studied prospectively between the period of October 2010 to June 2012 in the Department of Pathology, Kasturba Medical College Mangalore. Other causes of secondary thrombocytosis were excluded from the study. All the data for all the cases were obtained and were statistically analysed by Chi-square test and Fisher’s exact test. Results : One sixty cases were studied in the present study. Mean age in the present study was 43.64 years and the mean age of patients with malignancy was 53.18 years. Majority of the cases were benign (64.3%), followed by malignant (31.3%) and borderline tumors (4.4%). Mean platelet count in the study was 321x10 9 /L, and 80% of malignant cases had thrombocytosis as compared to 42.9% of borderline cases and 5.8% of benign cases, which is highly significant (p=0.0001). Predominant cases of ovarian carcinomas were in stage III (80%) followed by stage IV (17.5%) and stage II (2.5%). Conclusion : Pre-operative secondary thrombocytosis is a frequent finding in ovarian carcinoma and is significantly associated with advanced FIGO stage. The presence of thrombocytosis acts as a poor prognosticator in epithelial ovarian carcinoma.


Introduction
Thrombocytosis refers to platelet count above the normal value (> 400 x 10 9 /L). Thrombocytosis is classified according to its origin intoprimary and secondary types [1,2]. Primary thrombocytosis refers to persistent elevation of platelet count due to clonal thrombopoiesis as found in chronic myeloproliferative and some myelodysplastic disorder [1,2]. Secondary thrombocytosis can be due to a variety of underlying conditions like acute inflammation, malignancy, postsplenectomy and others. Short-lived secondary thrombocytosis is observed in conditions such as trauma, acute bleeding or major surgical procedures [2,3]. Secondary thrombocytosis associated with Malignancy may persist for a longer time [2,4]. Malignancy is one of the most important causes of secondary thrombocytosis [2,5]. The association between thrombocytosis and malignancies has been well demonstrated since many years [6]. One third of all cancer patients demonstrate thrombocytosis at the time of diagnosis [7].

Original Research Article
Pathology Update: Tropical Journal of Pathology & Microbiology Available online at: www.pathologyreview.in 282 | P a g e In ovariancancer, thrombocytosisisa poor prognostic factor in locally advanced disease [5,[21][22][23][24]. Chalas et al found thrombocytosis in 56% of the reviewed cases of epithelial ovarian malignancies.
Moreover, in the same study thrombocytosis was shown to have a predictive value of 83% and a specificity of 84% in differentiating malignant tumors from benign pelvic masses [25]. The etiology of neoplastic megakaryopoiesis remains unclear, it might be related to increased rate of platelet production [7,8]. Increased platelet production are due to certain cytokines (IL-6, IL-1) and growth factors released by malignant cells.
Platelet aggregates with tumor cells prevents immune mediated tumor cell clearance. Thrombospondin-1 helps in adhesion of circulating tumor cells to endothelium, extra vasation of tumor cells and metastasis. Elevated platelet counts may also have a role in ovarian cancer growth and metastasis [9].

Materials and Methods
In the present study, 160 cases of epithelial ovarian tumors were studied.
Type of study: Prospective study. The data were collected regularly during the study period from the case records at respective Medical Record Departments and laboratory reports at Kasturba Medical College Hospital, Ambedkar Circle.
All the data for all the cases were obtained.
2) Surgical and histopathological evidences of epithelial ovarian tumor.

5) Pre-operative chemotherapy
Sampling methods: The data collected in the present study were pre-operative platelet counts, stage (FIGO staging- Table 1), tumor histology and, the presence and degree of ascites among all the patients.
All patients underwent staging laparotomy. The gross specimens obtained after surgery were examined in detail. Tissue was fixed in 10% buffered formalin, and processed by paraffin embedding.
The blocks were serially cut, each of 3-5µ thickness and the sections counterstained with Haemotoxylin and Eosin (H&E).
The routine H and E stained slides of all the cases were reviewed thoroughly and the histopathological findings were recorded.
As per the latest WHO classification, the epithelial ovarian tumors were divided histologicaly into benign, borderline and malignant.

Original Research Article
Pathology Update: Tropical Journal of Pathology & Microbiology Available online at: www.pathologyreview.in 283 | P a g e

1C
The tumour is limited to one or both ovaries, with any of the following: • The ovary capsule is ruptured • The tumour is detected on the ovary surface • Positive malignant cells are detected in the ascites or peritoneal washings

2A
• The tumour has extended and/or implanted into the uterus and/or the fallopian tubes.
• Malignant cells are not detected in ascites or peritoneal washings

2B
• The tumour has extended to another organ in the pelvis • Malignant cells are not detected in ascites or peritoneal washings

2C
• Tumours are as defined in 2A/B, and malignant cells are detected in the ascites or peritoneal washings Stage 3-The tumour involves one or both ovaries with microscopically confirmed peritoneal metastasis outside the pelvis and/or regional lymph node metastasis.
Includes liver capsule metastasis.

3A
• Microscopic peritoneal metastasis beyond the pelvis

3B
• Microscopic peritoneal metastasis beyond the pelvis 2 cm or less in greatest dimension

3C
• Microscopic peritoneal metastasis beyond the pelvis more than 2 cm in greatest dimension and/or regional lymph nodes metastasis Stage: 4 -Distant metastasis beyond the peritoneal cavity. And, liver parenchymal metastasis.
Statistical methods: The incidence of thrombocytosis in the studied tumors were statistically analysed by Chi-square test. The patients with thrombocytosis were correlated with the FIGO stage and, presence and degree of ascites among malignant cases were statistically analysed by Chi-square test and Fisher's exact test. In the present study, a p value <0.05 was considered significant for the performed tests.

Results
In the present study, 160 cases of surface epithelial ovarian tumors (SEOT) were reviewed thoroughly. In an attempt to study the incidence of thrombocytosis among surface epithelial ovarian tumors, these tumors were classified into benign, borderline and malignant according to the recent WHO classification 2008 [26].
The mean age of all patients with SEOT was 43.64 years (median 42.00 years) as shown in Table 2. The mean age of patients for benign cases was 38.85 years. The mean age for borderline cases was 46 years, whereas the malignant had a mean age of 53.18 years. In the present study, out of 160 cases of surface epithelial ovarian tumors, majority of the cases were benign (64.3%), followed by malignant (50%) and borderline tumor cases (4.4%) as depicted in Table 3.  In the present study, serous ovarian tumors (70.6%) were the most common histologic type, followed by mucinous (27.5%) tumors. 49 (30.6%) cases of surface epithelial ovarian tumors had thrombocytosis. The platelet count ranged from 172x10 9 /L to 641x10 9 /L and mean the value was 321x10 9 /L. In this study 80% of malignant cases had preoperative thrombocytosis, as compared to 42.9% of borderline cases and 5.8% of benign cases, which is highly significant (p=0.0001) ( Table 4). 43 (26.9%) cases had ascites. The presence and degree of ascites was more in malignant SEOTs, than in borderline and benign tumors, and this finding was statistically highly significant (p=0.001). Ascites was moderate to massive in 42% and 36% cases, respectively of malignant SEOTs. Minimal ascites was present in 14% cases of borderline SEOTs. None of the benign SEOTs had ascites (Fig 1).  Thirty nine out of forty cases of malignant ovarian tumors with thrombocytosis were in advanced stage of disease (stage III /stage IV) and 1 (2.5%) case which was in stage II disease. Thirtytwo (80%) cases were cases were in stage IV. This correlation of thrombocytosis in malignant ovarian tumors with advanced FIGO stage was statistically highly significant (p=0.0001) ( Table 5).

Discussion
Pre-operative thrombocytosis is associated wit malignancies including ovarian neoplasia. According to recent WHO classification, the ovarian tumors have been classified into benign, borderline and malignant tumors which are further classified into their histologic subtypes. One hundred and sixty cases of surface epithelial ovarian tumors were included in our study. Majority of the study cases were benign (64.43%) followed by malignant (31.3%) and borderline tumors. The correlation of thrombocytosis with presence and degree of ascites, and FIGO staging were analyzed among malignant ovarian tumors.  [5,23]. In the present study, the mean platelet count was 321X10 9 /L and the platelet count ranged from 172X10 9 /L to 641X10 mean platelet counts for malignant epithelial ovarian tumors (ovarian carcinoma) was 435X10 /L to 641X10 9 /L) similar to the study done by Levin and Conley ( > 400X10 9 /L) [4].

1: Distribution of ascites among surface epithelial ovarian tumors
majority of surface epithelial ovarian carcinomas were in stage III (80%) followed by stage IV (17.5%). The borderline epithelial tumors were predominantly stage II (2.5%) diseases. This finding was statistically highly significant Thirty nine out of forty cases of malignant ovarian tumors with thrombocytosis were in advanced stage of disease (stage III /stage IV) and 1 (2.5%) case which was in stage II disease. Thirtytwo (80%) cases were cases were in stage IV. This correlation of thrombocytosis in malignant ovarian tumors with advanced FIGO stage was statistically highly significant (p=0.0001) ( Table 5).
operative thrombocytosis is associated with many malignancies including ovarian neoplasia. According to recent WHO classification, the ovarian tumors have been classified into benign, borderline and malignant into their histologic cases of surface epithelial ovarian tumors were included in our study. Majority of the study cases were benign (64.43%) ) and borderline (4.4%) The correlation of thrombocytosis with FIGO staging were analyzed among malignant ovarian tumors.
The mean age in the present study was 43.64 years (17-80 years) which is lower than those reported by other The mean age for malignant cases was 53.18 similar to the study done by Age recorded in our study was lower than the studies done by Li et al, Soonthornthum Obermair In the present study, /L and the platelet /L to 641X10 9 /L. The ts for malignant epithelial ovarian tumors (ovarian carcinoma) was 435X10 9 /L (226X10 9 /L) similar to the study done by Levin /L) [4]. Count in the present study was lower than the studies done by Crasta et al, Li et al and Bozkurt et al(610X10 446X10 9 /Lrespectively) [5,7,29]. But the count was higher than the results conducted by Kerpsack and Finan (298X10 9 /L) [24]. Forty (80%) out of fifty cases of malignant surface epithelial ovarian tumors had pre operative thrombocytosis as compared to only three (42.9%) cases of borderline tumors and six (5.8%) cases of benign tumors. These findings were statistically significant (p=0.0001) and higher than other studies done by Levin [5,7,29]. But the count was higher than the results conducted by Kerpsack and Forty (80%) out of fifty cases of malignant surface epithelial ovarian tumors had preperative thrombocytosis as compared to only three (42.9%) cases of borderline tumors and six (5.8%) cases of benign tumors. These findings were statistically significant (p=0.0001) and higher than other studies done by Levin [4-5,7,9,21-In the present study, ascites was present in 44 cases (27.5%), out of 160 cases. Ascites was more prevalent in malignant epithelial ovarian tumors (86%) as compared to borderline tumors (14%), which was higher than the studies done by Shen-Gunther and Mannel, and Puiffe et al (42% and 33%) [32,33]. Among malignant ovarian tumors with ascites, majority of the cases had moderate (42%) to massive ascites (36%) followed by minimal ascites seen only in 8% cases. All the borderline cases had minimal ascites and it was absent in all the benign ovarian tumors. In our study, statistical significance was associated with the presence or absence of ascites with FIGO staging (p=0.0001), which is similar to the results seen in Bouanene et al (p=0.020) [27]. In this study, majority of the ovarian carcinomas were in advanced stages (stage III-72% and stage IV-14%) followed by 10% in stage I and 4% were in stage II disease. This finding was statistically highly significant (p=0.0001). This finding was similar to the studies conducted by Chalas [5,7,9,21,23,[34][35].

Conclusion
Pre-operative secondary thrombocytosis is a frequent finding in malignant epithelial ovarian tumor (ovarian carcinoma) and is significantly associated with the presence and degree of ascites as well as advanced FIGO stage. This implies that pre-operative thrombocytosis is probably a marker of tumor aggressiveness, and that platelet may have a role in cancer growth and progression. Thus, the presence of pre-operative thrombocytosis has significance as a poor prognosticator in ovarian carcinoma.