Introduction
Hepatoblastoma is composed of various differentiated epithelial elements resembling fetal hepatocyte with or without mesenchymal elements. Hepatoblastoma is the most common hepatic malignancy in children [1]. In recent decades, diagnosis and treatment have been significantly improved in pediatric hepatoblastoma, and many patients may expect a long survival [1,2,3]. In contrast, adult hepatoblastoma is a very rare and aggressive neoplasm with a dismal prognosis. To date, only a few cases have been adequately reported in the medical literature, and no effective diagnostic and therapeutic strategy has been established. The aim of this study was to review all the case reports on adult hepatoblastoma, to explore the clinical features and outcomes, and to provide clinical clues for further study.
Methods
A PubMed search of English language reports from 1958 to 2011 was performed using the key words: ‘hepatoblastoma’ and ‘adult’. Articles were further considered after reviewing the titles and abstracts when available. Only those reports were included in which the diagnosis of hepatoblastoma was confirmed histologically by an autopsy, a biopsy or the resection specimen. Demographic information, clinical symptoms and signs, laboratory test and medical imaging, treatment and clinical outcomes were extracted and analyzed. Finally, in order to better characterize the outcome of adult hepatoblastoma, survival was assessed by the Kaplan-Meier method and group comparisons were performed using the log-rank test. A p value <0.05 indicated statistical significance. Analyses were conducted using SPSS 15.0 (SPSS, Chicago, Ill., USA).
Results
Case Report Overview
An extensive English literature search revealed only 36 reports including 40 patients with adult hepatoblastoma dating back to 1958 [4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40]. Before Carter [9] initially reported a patient with adult hepatoblastoma in the English literature, hepatic embryonic mixed tumor, mixed tumor and embryonic tumor had been reported as hepatoblastoma [5,6,7,8,10]. Eleven cases were reported before the 1980s, 20 cases were reported in the 1980s and the remaining 9 cases were reported in the present century. Of these, 16 cases were reported by authors from Asia (Japan, Korea, China, and India) and 12 cases by European authors (Italy, Germany, UK, France, Sweden, and the Slovak Republic); 9 cases were reported by US authors, and the other 3 cases were reported by authors from Australia, Mexico and South Africa, respectively.
Baseline Characteristics
The male:female ratio was 1.1 (21:19). The mean age at diagnosis was 41.5 years, and the age span was 19–84 years (fig. 1). Most cases were in their 20s. Clinical symptoms were confirmed in 31/40 cases, 3/40 cases were incidental cases, and in 6/40 cases there were no available data in the literature. Pain was the most common symptom (23/31, 74.2%), including epigastric pain in 14 cases, right upper abdomen pain in 8 cases, abdominal pain in 3 cases, and lumbago in 1 case. Other clinical symptoms included fever, gastrointestinal symptoms, such as anorexia, nausea and vomiting, and rare findings, such as leg swelling, upper gastrointestinal bleeding and dyspnea. Twenty-five cases had positive findings on physical examination. The most common signs were as follows: abdominal mass (9/25, 36.0%), hepatomegaly (10/25, 40.0%) and weight loss (8/25, 32.0%). Other rare signs were leg edema, jaundice and anemia. Adult hepatoblastoma often presents with a giant mass, and almost all patients had a main tumor of more than 5 cm in diameter with a median size of 15 cm in diameter. Adult hepatoblastoma affects the right or left liver lobes, and the tumors are very often single (29/40, 72.5%). Patient characteristics were shown in table 1.
Diagnostic Modalities
In the literature, laboratory studies revealed nonspecific results. A slight increase in the serum hepatic zymogram was observed in a few cases. Of all 40 patients, only 6 cases tested positive for hepatitis B virus; the remaining 34 cases were negative hepatitis B virus, and in some cases, no descriptions of hepatitis status were found on careful reading. Fifteen of the patients were positive for α-fetoprotein (AFP), and the serum AFP level ranged widely from 35,819 to 1,548,000 ng/ml. In addition, serum vitamin K antagonist-II was elevated in 3 cases (case 32: 579,000 mAU/ml; case 3: 6,880 mAU/ml, and case 38: 38,496 mAU/ml).
The radiological investigations commonly used consisted of abdominal X-ray plain film, ultrasound, computed tomography, magnetic resonance imaging and selective celiac angiography. In earlier studies, X-ray film was the main imaging method; it showed elevation of the diaphragm, an enlarged liver and intratumoral calcification. After that, ultrasound, computed tomography, magnetic resonance imaging gradually became the most common imaging methods, and revealed clear liver lesions. Selective celiac angiography was performed in 18 cases and almost all patients (14/18, 77.8%) were characterized by high vascularization. Only 1 case (1/7, 14.3%) of those who underwent needle biopsy had a positive diagnosis of hepatoblastoma. Histopathologic examination was the final diagnostic method. Necrosis and hemorrhage were confirmed in 21 cases (52.5%) and calcification was found in 8 cases (20%). The diagnostic modalities are summarized in table 2.
Management Modalities and Outcome
Management modalities are shown in table 2. Seven patients did not receive any treatment, and the diagnosis of hepatoblastoma was made post mortem. Exploratory biopsy was performed in 8 cases, including 1 gastrojejunostomy patient, 2 peritoneoscopy patients and 1 case combined with systemic chemotherapy. Two patients received systemic chemotherapy or local transcatheter arterial chemoembolization. Twenty-three cases of adult hepatoblastoma underwent radical surgery, including surgical resection alone in 14 cases and comprehensive treatment in 9 patients. Comprehensive therapy modalities are shown in table 3.
Follow-up data were available in 27 patients. Overall outcome of the patient was poor since most of the patients died shortly after diagnosis and treatment. The overall median survival time was 4 months, and the 1-year survival rate was 29.6% (fig. 2). Younger patients showed significantly better prognoses than older patients (fig. 3). The median survival for patients younger than 50 years was 9 months with a 1-year survival rate of 41.2%; however, the median survival for patients older than 50 years was 2 months with a 1-year survival of 10.0% (χ2 = 4.128, p = 0.042). Patients who underwent radical treatment had improved survival compared to those who underwent biopsies and/or nonsurgical treatment (fig. 4). One-year survival for patients who underwent radical resection was 44.4% with a median survival of 10.5 months and 0% in those who did not undergo resection (χ2 = 17.488, p = 0.000). In Cox multivariate analysis, curative liver resection was an independent prognostic factor for improved survival (p = 0.003). The median survival for cases reported before 1995 and after 1995 was 2 and 6 months, respectively; however, no significant difference was shown (χ2 = 0.718, p = 0.397). The main tumor size also had no impact on survival (χ2 = 0.643, p = 0.422).
Fig. 2
Overall survival of adult hepatoblastoma (n = 27). Median survival: 4 months (95% confidence interval 0–12.35), 1-year survival: 29.6%.
Fig. 3
Overall survival of adult hepatoblastoma patients younger than 50 years (n = 17) and older than 50 years (n = 10). One-year survival for patients younger than 50 year: 41.2%; 1-year survival for patients older than 50 year: 10.0% (log rank test, χ2 = 4.128, p = 0.042).
Fig. 4
Overall survival of adult hepatoblastoma patients who underwent resection (n = 18) and those who did not (n = 9). One-year survival for patients who underwent resection: 44.4%; 1-year survival for patients who did not undergo resection: 0% (log rank test, χ2 = 17.488, p = 0.000).
Discussion
Hepatoblastoma may occur in association with genetic disorders. Parada et al. [26] cytogenetically investigated an adult patient with hepatoblastoma and found a genetic variation showing a hypertriploid stemline with multiple numerical and structural chromosomal aberrations. Weber et al. [41] screened 34 hepatoblastoma tumors and 3 hepatoblastoma cell lines for DNA copy number changes. The results identified gains on chromosomes 1q and 2 as the hallmark of DNA copy number changes in hepatoblastoma, 2q24 being a critical chromosomal band; this study also provided evidence that gains on 8q and 20 may play a role as markers of prognostic significance in hepatoblastoma. In addition, the Wnt signaling pathway may play an important role in the development of this malignant neoplasm [42]. Chronic hepatitis B and C virus infection and liver cirrhosis are causally associated with the development of hepatocellular carcinoma [43]. In the present review, there were only 6 patients definitely positive for hepatitis B virus. Liver cirrhosis was found in 7 patients. Interestingly, childhood hepatoblastoma classically arises within a healthy liver, unaffected by underlying disease. Whether there is a link between hepatoblastoma and hepatitis needs further investigation.
The most frequent symptoms of hepatoblastoma are abdominal pain, and the most frequent physical signs were abdominal mass, hepatomegaly and weight loss, but these signs are also found in other types of tumors. No obvious rule can be found regarding age distribution at diagnosis. Nor was there an obvious sexual difference in incidence in adult hepatoblastoma; however, this may be different from childhood hepatoblastoma [44]. Hepatoblastoma commonly forms a single giant mass within the liver. Laboratory data in cases of hepatoblastoma are not specific. Liver function tests were normal or mildly elevated. Serum AFP levels were elevated in 15 cases with a median value of 30,000 ng/ml. In some cases, the serum AFP level even exceeded 1,000,000 ng/ml [27,31]. These variations may be due to a difference in biological characters as in other tumors [4].
Early detection and diagnosis are believed to improve the prognosis of adult patients with hepatoblastoma. Therefore, it is important to be suspicious of hepatoblastoma when patients present with nonspecific symptoms. It should also be noted that liver function tests, hepatitis status and tumor markers, such as AFP, are not specific as mentioned previously. Hepatoblastoma is difficult to differentiate from other malignant tumors of the liver using imaging techniques. Kishimoto et al. [16] mentioned that intratumoral calcification demonstrated by computed tomography scans was somewhat helpful in the diagnosis of adult hepatoblastoma. The incidence of tumor calcification was significantly higher than in other liver tumors in mixed type [38]. Calcification may be characteristic in all liver tumors with mesenchymal elements [4]. In the present systematic study, we only confirmed 8 cases with calcifications. Morphological characteristics of adult hepatoblastoma are cystic changes and hypervascularity [4]. Cystic changes may be consecutive to intratumoral hemorrhage or necrosis, which is a common pathological finding. Arteriography could detect hypervascular tumors.
Harada et al. [21] insisted on the importance of needle aspiration biopsy to diagnose hepatoblastoma preoperatively. The present review does not consider this procedure to be adequate for a preoperative diagnosis because of its low positivity rate and the risk of bleeding and tumor dissemination. Needle biopsy was successful in only 1 of 7 cases. Postoperative pathological examination has proven to be the best approach for confirming hepatoblastoma. Ishak and Glunz [45] classified hepatoblastoma into two groups: an epithelial type, and a mixed epithelial and mesenchymal type. The epithelial type consists of fetal and embryonic cells. In this study, the mixed type was the most common histological type of adult hepatoblastoma.
From the present review of the literature, it can be seen that there is no standardized management of adult hepatoblastoma. Surgery was the first-line approach, but pre- and postoperative chemotherapy was administered in a few cases. Until now, there has been no established chemotherapy regimen. The chemotherapy protocols used in previous cases were platinum, Adriamycin, irinotecan and pirarubicin. Radical surgical excision appears to be the ‘gold standard’. Rougemont et al. [46] suggested treating adult hepatoblastoma in the same way as childhood hepatoblastoma: start treatment with cisplatin-based preoperative chemotherapy, followed by surgery.
Hepatic resection can considerably prolong survival and remains the primary treatment for adult hepatoblastoma. Many adult hepatoblastoma patients are unresectable at diagnosis and often develop intrahepatic recurrence and extrahepatic metastasis after liver resection. For these patients, adjuvant chemotherapy should be a conceivable choice in order to shrink the mass, thus facilitating its complete excision, and prolonging disease-free survival. Nakamura et al. [31 ]reported a patient with initially unresectable adult hepatoblastoma successfully treated by multimodal treatment, including intensive preoperative chemotherapy, liver resection and postoperative adjuvant chemotherapy who survived over 4 years after diagnosis. Ethanol injection and radiofrequency catheter ablation have been safe and effective in the treatment of primary or metastatic liver cancer. There was an adult hepatoblastoma patient who survived for 151 months after an initial operation with ethanol injection for her repeating recurrences [24]. Ke et al. [28] reported a patient who underwent surgical resection for a larger tumor nodule and radiofrequency catheter ablation for a smaller one. In addition, the role of chemoembolization as a rescue treatment for recurrences of resected hepatoblastoma was evaluated recently in a case report, and the patient survived 11 months after chemoembolization [37]. Multicenter efforts should focus on implementing standardized guidelines [46].
The prognosis of hepatoblastoma is extremely poor. The median survival was only 4 months with a 1-year survival of 29.6%. Younger patients had significantly better prognoses than older patients. Patients who underwent resection displayed improved survival compared to those who underwent biopsies and/or nonsurgical treatment. In Cox multivariate analysis, curative liver resection was an independent prognostic factor for improved survival. These findings are similar to those of a previous review [46]. Although no significant difference was shown, the median survival for cases reported after 1995 was better than that of cases reported before 1995, suggesting an improvement in the management of hepatoblastoma in adults [46]. Improvements in outcome will require the development of more effective systemic therapies.
Disclosure Statement
The authors have no potential conflict of interest to disclose.
Get Permission
PubMed ID
