CASE REPORT
Vol.42 No.08
Thrombopoietin-Producing Hepatocellular Carcinoma
Tomiko Ryu, Satoshi Nishimura, Hideaki Miura, Haruki Yamada, Haruhiko Morita*, Hiroshi Miyazaki*, Shigehiro Kitamura**, Yasusada Miura and Toshikazu Saito
A 66-year-old man with hepatocellular carcinoma (HCC) showed marked thrombocytosis (110.7×104/μl). Bone marrow (BM) aspirates demonstrated an increase of mature megakaryocytes (MgK). The serum thrombopoietin (TPO) level was increased to about 100-fold that of the normal level in the terminal stage. However, the platelet count gradually decreased to 13.5×104/μl. The autopsy specimen revealed normoplastic BM with decreased MgK, mainly consisting of the immature type, and it was negative for tumor cells. Liver specimen showed markedly fatty metamorphosis. Immunohistochemical staining of TPO demonstrated that hepatocytes were weakly stained and HCC cells strongly stained, suggesting TPO-producing HCC.
(Internal Medicine 42: 730–734, 2003)
Key words: thrombopoietin, thrombopoietin-producing tumor, hepatocellular carcinoma, thrombocytopoiesis
Introduction
Thrombopoietin (TPO), the primary hematopoietic growth factor involved in the regulation of megakaryocyte (MgK) growth and development as well as platelet production has recently been isolated and cloned (1–4). Although the complexities of TPO regulation are not fully understood, two possible mechanisms have been postulated. One mechanism is that TPO plasma levels are dependent on the rate of platelet/MgK TPO receptor (Mpl)-mediated uptake and catabolism (5–7). An elevated platelet level would lead to increased binding and catabolism of TPO to platelet/MgK receptors and consequently a decreased plasma TPO level, thereby limiting MgK production which would lead to a decreased production of platelets. The other mechanism is that plasma TPO concentrations are probably regulated by feedback control at the level of gene expression. McCarty et al (8) showed that TPO mRNA expression in the bone marrow and spleen is increased in thrombocytopenic animal models. Sungaran et al (9) demonstrated that in thrombocytopenic human subjects with aplastic anemia, postchemotherapy marrow aplasia, and immune thrombocytopenia, the stromal cells show marked TPO mRNA expression, suggesting that TPO mRNA expression in human bone marrow might be modulated by platelet mass. However, these two mechanisms may not be mutually exclusive or operate in concert.
To date hepatoblastoma and ovarian carcinoma have been reported to produce TPO. Nickerson et al (10) described two children with hepatoblastoma, marked thrombocytosis, and extremely elevated alpha fetoproteins. They suggested that a specific cytokine like TPO affecting megakaryocytopoiesis may be associated with excessive MgK seen in the bone marrow and consequent platelet production. Furuhashi et al (11) reported a case of TPO-producing ovarian carcinoma confirmed by immunohistochemistry. They attempted to study the serum TPO levels because of marked thrombocytosis.
Here, we describe a case of TPO-producing hepatocellular carcinoma (HCC) with thrombocytosis, and discuss the causes of thrombocytopenia observed with progression of HCC in the terminal stage.
Case Report
A 66-year-old man had symptoms of abdominal distension and appetite loss two months before admission in April 2000. He had a drinking habit. Physical examination showed a huge hard mass in the abdomen. Laboratory data showed leuko- and thrombocytosis, that is, the white blood cell (WBC) count was 12,800/μl, the differential showed 55% neutrophils and 38% lymphocytes and the platelet count was 67.4×104/μl. Asparate aminotransferase (AST), and alanine aminotransferase (ALT) were 63 and 19 IU/l, respectively. Prothrombin time and activated partial thromboplastin time were in the normal range. Alpha fetoprotein was 1.7 ng/ml. Hepatitis B surface antigen, and anti-hepatitis C antibody were negative. Ultrasonography (US) and computed tomography (CT) (Fig. 1) of the abdomen demonstrated multiple liver tumors and the largest one measured 15×11×18 cm in the left lobe. Hepatocellular carcinoma (HCC) was diagnosed by needle biopsy and the background liver demonstrated chronic active hepatitis (non-B, non-C). After he was treated by transcatheter arterial embolization (TAE) of the left hepatic artery, the platelet count was decreased from 67.4×104/μl to 48.0×104/μl. As an enlargement of HCC was not seen, he was discharged in August 2000. He was re-admitted because of abdominal pain in January 2001. As the platelet count increased to 86.7×104/μl, bone marrow aspiration was performed. All nucleated bone marrow cells were 12.0×104/μl and megakaryocyte (MgK) count was 125/μl. Bone marrow specimen showed a marked increase in mature, platelet-producing MgK, negative for tumor cells (Fig. 2). Platelet aggregation studies to ADP were normal. Among the several kinds of cytokines affecting megakaryocytopoiesis, the serum levels of thrombopoietin (TPO) (12), erythropoietin (EPO), interleukin (IL)-6, transforming growth factor (TGF)-β1 and hepatocyte growth factor (HGF) were increased. In particular, the serum TPO level was extremely high, 14.73 fmol/ml (normal range 0.40±0.29). With the progression of HCC, the serum TPO level continuously increased. However, the platelet count was inversely proportional to serum TPO level. The platelet count gradually decreased to 13.5×104/μl, despite the increased TPO level, of 38.45 fmol/ml in August (Fig. 3). Thrombocytopenia caused by disseminated intravascular coagulation (DIC) was denied. Like TPO, the serum levels of EPO, IL-3 and IL-6 were increased in August, compared with those in March (Table 1). The patient died in September 2001. Immunohistochemical staining of TPO was performed. A 4-μm section was cut from the paraffin blocks of liver. Each section was mounted on a glass slide and deparaffinazed. Rabbit anti-human TPO polyclonal antibody, diluted 200 times (a gift from Kirin Brewery Co., Ltd., Tokyo, Japan) was applied for 2 hours at 37ºC. The primary antibody was visualized using the Simple Stain MAX-PO (Nichirei, Tokyo, Japan) according to the instruction manual (13). Although hepatocytes with fatty metamorphosis were weakly stained, HCC cells were clearly stained by anti-TPO antibody, suggesting TPO-producing HCC (Figs. 4, 5, 6). As a control, immunohistochemical staining of TPO was performed using liver tissues of another patient with HCC and liver cirrhosis. HCC cells were negative and hepatocytes were weakly positive (Figs. 7, 8). BM metastasis, apparent splenomegaly and thrombosis due to DIC were not observed.
Discussion
We described a 66-year-old man with HCC and marked thrombocytosis, which is very rare among HCC patients. After he was treated by TAE of the left hepatic artery, the platelet count decreased from 67.4×104/μl to 48.0×104/μl, and then recovered. Although the serum TPO level was not measured, the decreased platelet count could be related to the decreased TPO level based on the reduction in mass volume. For examination of thrombocytosis (platelet count 110.7×104/μl), bone marrow aspirates were obtained. The bone marrow specimen showed a marked increase in mature, platelet-producing MgK, negative for tumor cells. Several kinds of cytokines affecting megakaryocyto- and thrombocytopoiesis were examined, and a high serum TPO level was noted. With the progression of HCC, the serum TPO level continuously increased. Immunohistochemical staining of TPO was performed using an autopsy liver specimen. Although hepatocytes with fatty metamorphosis were weakly positive, HCC cells were clearly stained by anti-TPO antibody, suggesting TPO-producing HCC (Figs. 4, 5, 6). In situ hybridization of the tumor material to complimentary DNA of TPO was not examined.
Table 2 shows platelet counts and plasma levels of endogenous TPO in thrombocytopenic patients (14). The encoded polypeptides of TPO have a predicted molecular mass of approximately 35,000 kDa. If TPO consists of a full-length form, it is possible to convert fmol/ml to pg/ml (2 fmol/ml×35,000=70 pg/ml). However, if the TPO molecule was truncated by proteolysis, the conversion was not correct.
With the progression of HCC, the platelet count was inversely proportional to the serum TPO level. The autopsy specimen showed normoplastic BM with decreased MgK, mainly consisting of immature type (Fig. 9). The causes of thrombocytopenia in the terminal stage could be attributed to: ① TPO dysfunction, ② an increase of cytokines affecting thrombocytopoiesis, or ③ disappearance of platelet TPO receptor (Mpl) via catabolism.
① Anti-TPO antibody neutralizing TPO activity was measured with a radioimmunoassay using glycosylated recombinant full-length human TPO as the antigen, as previously described (15) and the result was negative. The erythropoietin (EPO)-like domain of TPO contains the receptor-binding portion which is biologically active. Even though the C-terminal domain was detached from the full-length, 70-kDa form by proteolytic conversion, it is still possible to detect TPO protein by TPO/ELISA, even with the loss of biological activity. ② Among cytokines affecting thrombocytopoiesis, the serum levels of cytokines inhibiting megakaryocytopoiesis (16, 17), such as TGF-β1, platelet factor 4 (PF4), β-thromboglobulin (β-TG) were increased against megakaryocytopoiesis accelerated by TPO. However, there was no change in serum cytokine levels in March and September. In contrast, the serum levels of EPO, IL-3 and IL-6 stimulating megakaryocytopoiesis (18, 19) were enhanced in August, compared with that in March. Suppressed megakaryocytopoiesis may have induced these cytokines, or there is a possibility that HCC cells produced multi-cytokines, although immunohistochemical staining, other than TPO, was not performed. A physiological regulator enhancing TPO gene expression, and hepatocyte growth factor (HGF) (20) showed a rather high level (Table 1). ③ Cohen-Solal K et al (21) examined whether or not, the expression of Mpl is regulated by its ligand, TPO. The levels of Mpl transcripts and Western blot analysis of MgK produced in the presence or absence of TPO, showed no difference in Mpl levels. These results indicated that TPO does not have a major effect on the transcription or translation of Mpl. However, they also suggested that an excess of circulating TPO could lead to the disappearance of Mpl from platelets via catabolism. As TPO production was self-supported by HCC in this case, the serum TPO level was extremely high. Consequently, it may mimic down-regulation of Mpl.

Acknowldgements: We would like to thank Asako Takematsu, Kazuko Mochizuki, Kazuhiko Kobayashi, and Chiemi Kurita for their excellent technical assistance.
References
1)Bartley TD, Bogenberger J, Hunt P, et al. Identification and cloning of a karyocyte growth and development factor that is a ligand for the cytokine receptor. Mpl. Cell 77: 1117–1124, 1994.
2)de Sauvage FJ, Hass PE, Spencer SD, et al. Stimulation of megakaryocytopoiesis and thrombopoiesis by the c-Mpl ligand. Nature 369: 533–538, 1994.
3)Lok S, Kaushansky K, Holly RD, et al. Cloning and expression of murine thrombopoietin cDNA and stimulation of platelet production in vivo. Nature 369: 565–568, 1994.
4)Ogami K, Shimada Y, Sohma Y, et al. The sequence of a rat cDNA encoding thrombopoietin. Gene 158: 309–310, 1995.
5)Kuter DJ, Rosenberg RD. The reciprocal relationship of thrombopoietin (c-Mpl ligand) to changes in the platelet mass during busulfan-induced thrombocytopenia in the rabbit. Blood 85: 2720–2730, 1995.
6)Stoffel R, Wiestner A, Skoda RC. Thrombopoietin in thrombocytopenic mice: evidence against regulation at the mRNA level and for a direct regulatory role of platelets. Blood 87: 567–573, 1996.
7)Nagata Y, Shozaki Y, Nagahisa H, Nagasawa T, Abe T, Todokoro K. Serum thrombopoietin level is not regulated by transcription but by the total counts of both megakaryocytes and platelets during thrombocytopenia and thrombocytosis. Thromb Haemost 77: 808–814, 1997.
8)McCarty JM, Sprugel KH, Fox NE, Sabath DE, Kaushansky K. Murine thrombopoietin mRNA levels are modulated by platelet count. Blood 86: 3668–3675, 1995.
9)Sungaran R, Markovic B, Chong BH. Localization and regulation of thrombopoietin mRNA expression in human kidney, liver, bone marrow, and spleen using in situ hybridization. Blood 89: 101–107, 1997.
10)Nickerson HJ, Silberman TL, McDonald TP. Hepatoblastoma, thrombocytosis, and increased thrombopoietin. Cancer 45: 315–317, 1980.
11)Furuhashi M, Miyabe Y, Oda H. A case of thrombopoietin-producing ovarian carcinoma confirmed by immunohistochemistry. Gynecol Oncol 74: 278–281, 1999.
12)Tahara T, Usuki K, Sato H, et al. A sensitive sandwich ELISA for measuring thrombopoietin in human serum: serum thrombopoietin levels in healthy volunteers and in patients with haemopoietic disorders. Br J Haematol 93: 783–788, 1996.
13)Kanamori Y, Kigawa J, Itamochi H, et al. Correlation between loss of PTEN expression and Akt phosphorylation in endometrial carcinoma. Clin Cancer Res 7: 892–895, 2001.
14)Kuter DJ, Hunt P, Sheridan W, Zucker-Franklin D. Thrombopoiesis and Thrombopoietins. Molecular, Cellular, Preclinical, and Clinical Biology. Humana Press, Totowa, New Jersey, 1997: 359–375.
15)Shiozaki H, Miyawaki S, Kuwaki T, Hagiwara T, Kato T, Miyazaki H. Autoantibodies neutralizing thrombopoietin in a patient with amegakaryocytic thrombocytopenic purpura. Blood 95: 2187–2188, 2000.
16)Ishibashi T, Miller SL, Burstein SA. Type beta transforming growth factor is a potent inhibitor of murine megakaryocytopoiesis in vitro. Blood 69: 1737–1741, 1987.
17)Han ZC, Sensebe L, Abgrall JF, Briere J. Platelet factor 4 inhibits human megakaryocytopoiesis in vitro. Blood 75: 1234–1239, 1990.
18)Hill RJ, Warren MK, Levin J, Gauldie J. Evidence that interleukin-6 does not play a role in the stimulation of platelet production after induction of acute thrombocytopenia. Blood 80: 346–351, 1992.
19)Debili N, Masse JM, Katz A, Guichard J, Breton-Goriu J, Vainchenker W. Effects of the recombinant hematopoietic growth factors interleukin-3, interleukin-6, stem cell factor, and leukemia inhibitory factor on the megakaryocytic differentiation of CD34+ cells. Blood 82: 84–95, 1993.
20)Yamashita K, Matsuoka H, Ochiali T, et al. Hepatocyte growth factor/scatter factor enhances the thrombopoietin mRNA expression in rat hepatocytes and cirrhotic rat livers. J Gastroenterol Hepatol 15: 83–90, 2000.
21)Cohen-Solal K, Vitrat N, Titeux M, Vaincheuker W, Wendling F. High-level expression of Mpl in platelets and megakaryocytes is independent of thrombopoietin. Blood 93: 2859–2866, 1999.


From the Department of Internal Medicine, ** Department of Pathology, Social Insurance Chuo General Hospital, Tokyo and *Pharmaceutical Research Laboratory, Kirin Brewery Co. Ltd., Takasaki, Gunma
Received for publication August 9, 2002; Accepted for publication March 14, 2003
Reprint requests should be addressed to Dr. Tomiko Ryu, the Department of Internal Medicine, Social Insurance Chuo General Hospital, 3-22-1 Hyakunin-cho, Shinjuku-ku, Tokyo 169-0073
go to JSIM
Copyright(C) 1997-2004, The Japanese Society of Internal Medicine. Allright reserved.
E-mail : naika@naika.or.jp Last Up Date 2003/8/25