Thrombopoietin in Patients With Hepatoblastoma

To the Editor:

Recently the major physiological regulator of the thrombopoiesis has been purified and termed thrombopoietin (TPO), c-mpl ligand. Using in situ hybridization technique, Sungaran et al1 found strong TPO mRNA expression in the hepatocytes in the human liver which could be the major source of TPO production. On the other hand, marked thrombocytosis (> 50 × 10⁴/μL) is frequently seen in patients with hepatoblastoma.2 Therefore, we assessed serum TPO levels and looked into its biological as well as clinical significance in hepatoblastoma. Schweinitz et al3 suggested interleukin-6 (IL-6) to be one of the mediators, showing that hepatoblastoma cells induce IL-6 production in surrounding fibroblasts and endothelial cells by virtue of the endogenous secretion of interleukin-1β (IL-1β). On the other hand, TPO has been shown to be produced mainly in the liver.4 Therefore, serum TPO levels were assessed in seven untreated patients with hepatoblastoma by using a sandwich enzyme-linked immunosorbent assay (ELISA)5 in addition to IL-1β and IL-6 levels. High serum TPO levels were observed in all samples examined, ranging from 3.15 to 11.02 fmol/mL (mean ± SD; 6.08 ± 1.25, normal range is 0.33 to 1.72 in adults,5 but not currently available in childhood) (Table 1). Serum IL-1β and IL-6 levels (mean ± SD) in five patients were 0.32 ± 0.08 and 22.9 ± 7.9 pg/mL, respectively (Table 1). IL-6 levels were also higher than normal range. Platelet counts appeared to correlate relatively with serum TPO levels (P = .1), but not with IL-1β and IL-6 levels. Additionally, a serial measurement in a patient indicated that TPO levels decreased along with the normalization of platelet counts during sequential chemotherapy. Before the first through the fifth chemotherapy, TPO levels were 10.1, 7.11, 1.09, 3.90, and 1.75 fmol/mL, while platelet counts were 115.0, 77.2, 66.5, 35.5, and 39.4 × 10⁴/μL, respectively. Furthermore, we found the expression of c-mpl mRNA in hepatoblastoma tissues (five of eight cases), suggesting that TPO acts in a way of autocrine and/or paracrine system for hepatoblastoma cells. From these observations, we came to the conclusion that thrombocytosis in hepatoblastoma exclusively resulted from the production of TPO within tumor tissues, and further that TPO might be a growth factor for hepatoblastoma. Thus, clinical application of recombinant TPO in hepatoblastoma would better be concerned with its possible potential to promote tumor progression.