Objectives: Our program routinely used fluorodeoxyglucose–positron emission tomography/computed tomography as part of the liver transplant evaluation of patients with hepatocellular carcinoma. The aim of this study was to evaluate the role of this imaging modality in the pretransplant work-up.
Materials and Methods: This was a retrospective chart review of our liver transplant database from January 2011 to December 2014 for all patients with hepatocellular carcinoma who underwent a liver transplant. Collected data included age, sex, cause of liver disease, imaging modality, fluorodeoxyglucose–positron emission tomography/computed tomography results, explant tissue analysis, type of transplant, and transplant outcome.
Results: During the study period, 275 liver transplants were performed. Fifty-three patients had hepatocellular carcinoma; 41 underwent fluorodeoxyglucose–positron emission tomography/computed tomography. Twenty-nine patients underwent living-donor liver transplant, and 12 patients underwent deceased-donor liver transplant. One of the 41 patients with negative FDG-imaging results had no evidence of hepatocellular carcinoma in the explant and was excluded from the study. The patients’ average age was 58 years (range, 22-72 y), and 28 patients were men. The cause of liver disease was hepatitis C virus in 24 patients, cryptogenic cirrhosis in 12 patients, and hepatitis B virus in 5 patients. One patient had no hepatocellular carcinoma on explants and was excluded from the study. Twenty-five patients had hepatocellular carcinoma that met the Milan criteria, 7 were within the UCSF (University of California, San Francisco) criteria, and 8 exceeded the UCSF criteria. Of the 40 patients, 11 had positive fluorodeoxyglucose–positron emission tomography/computed tomography results (27.5%) with evidence of hepatocellular carcinoma in the explant; the remaining 29 patients (72.5%) had negative results. The fluorodeoxyglucose–positron emission tomography/computed tomography results were positive in 16% (4 of 21) of patients who met the Milan criteria, 28% (2 of 7) of patients who met the UCSF criteria and 62% (5 of 8) of patients who exceeded the UCSF criteria.
Conclusions: Fluorodeoxyglucose–positron emission tomography/computed tomography has a low degree of use in patients with hepatocellular carcinoma that falls within the Milan criteria and should not be routinely used as part of the liver transplant work-up.
Key words : Cancer diagnosis, FDG-PET CT, Hepatocellular carcinoma, Liver transplant, Magnetic resonance imaging, Radiography
Hepatocellular carcinoma (HCC) is the third most common cause of cancer-related deaths worldwide,1 and the burden of this devastating disease is expected to increase. Each year, approximately 750 000 new cases are diagnosed. Risk factors include preexisting infection with hepatitis B virus and cirrhosis.2-5 Hepatitis C carcinoma is one of the most common cancers in parts of Africa and the Far East.6 In developed countries, incidence peaks between the ages of 40 and 60 years, but in developing countries where the disease is common, most cases occur between ages 20 and 40 years.
The diagnosis of HCC in our study was made based on radiologic criteria that followed the American Association for the Study of Liver Diseases guidelines.7,8 Positron emission tomography (PET) is a noninvasive modality for whole-body imaging of cellular and metabolic functions and has been used in some patients. The concept of PET was first conceived of in the early 1970s. Cancer imaging with fluorodeoxyglucose (FDG)–PET, which was first described by Otto Warburg in the 1931, takes advantage of increased glucose metabolism in cancer cells.9
Cells take up FDG along with glucose transporters such as glucose transporter-1. Once in the cytoplasm, FDG enters the first step of the glycolytic pathway and is converted to FDG-6-phosphate. Unlike glucose-6-phosphate, FDG-6-phosphate is not a substrate of phosphoglucose isomerase, the second enzyme in the glycolytic pathway. Hence, the radioactivity accumulates inside cells. Selective imaging of cancer cells is practical because of persistent glucose uptake by many cancer cells in a low-serum insulin state, when most cells outside the central nervous system consume glycogen or fatty acid instead of glucose for energy needs. This differential cellular metabolism produces a high contrast between cancer cells and surrounding tissue in a fasting patient. This “contrast resolution,” in turn, makes FDG-PET unique among imaging modalities, including advanced radiography, computed tomography (CT) and magnetic resonance imaging.10 The sensitivity of 18F-FDG PET for detecting HCC is not superior to that of conventional imaging (50%-70%)11-15 because well-differentiated HCC has a high rate of gluconeogenesis comparable with that of normal liver tissue, resulting in similar uptake of 18F-FDG. In contrast, high diagnostic performance has been reported for 18F-FDG for the detection of other primary liver malignancies, such as cholangiocarcinoma and hepatocholangiocarcinoma or liver metastases.
Choline is one of the components of phosphatidylcholine, which is an essential element of phospholipids in the cell membrane. As a result of the higher choline content of cells in HCC compared with normal liver tissue,16,17 it can be detected using magnetic resonance spectroscopy. In comparing the 2 PET CT modalities, it became clear that 18F-fluorocholine was better than 18F-FDG for detecting HCC; further, a trend toward more intense 18F-fluorocholine uptake was seen in well-differentiated HCC cells compared with poorly differentiated HCC cells.18 Our program uses FDG-PET CT routinely in the pretransplant evaluation of patients with HCC. The aim of our study was to evaluate the utility of FDG-PET CT in the setting of the pretransplant liver work-up in HCC patients.
Materials and Methods
This was a retrospective chart review of our liver transplant (LT) database of all patients with HCC who underwent a FDG-PET CT scan before LT from January 2011 to December 2014. During this period, 275 LTs were performed: 183 living-donor LTs (LDLTs) and 92 deceased-donor LTs (DDLTs). Fifty-three patients had been diagnosed with HCC, and FDG-PET CT was performed on 41 patients. The data collected included the following: age, sex, cause of liver disease, tumor size, and number as seen on CT or magnetic resonance imaging, PET CT results, explant tissue analysis, type of transplant (LDLT or DDLT) and transplant outcome. Patients whose HCC met the Milan criteria received a Model for End-Stage Liver Disease (MELD) score exception of 22 on the DDLT waiting list, patients who met the UCSF (University of California, San Francisco) criteria were allowed to undergo LDLT, and patients who exceeded the UCSF criteria received an LT if their HCC was successfully downstaged following locoregional therapy.
The study was approved by the office of research administration at our institution, patients' written informed consent was waived, and all the protocols conformed to the ethical guidelines of the 1975 Helsinki Declaration.
The average age of patients was 58 years (range, 22-72 y), and 28 patients were men. The cause of liver disease was hepatitis C virus in 24 patients, cryptogenic cirrhosis in 12 patients and hepatitis B virus in 5 patients. The median follow-up was 38 months. Twenty-five patients had HCC that met the Milan criteria, 7 patients met the UCSF criteria, and 8 patients exceeded the UCSF criteria. Twenty-nine patients underwent LDLT, and 12 patients underwent DDLT. One patient with negative FDG-PET CT results had no evidence of HCC in the explant and was excluded from the study. Of the 40 patients with HCC who underwent FDG-PET CT, 11 patients had a positive result (27%) with evidence of HCC in the explant: 4 whose HCC met the Milan criteria, 2 who met the UCSF criteria, and 5 who exceeded the UCSF criteria. Among the 11 patients, the explant showed well-differentiated HCC in 5 patients, moderately differentiated HCC in 3 patients, poorly differentiated HCC in 1 patient, and complete necrosis after locoregional therapy in 2 patients.
Of all the transplant recipients, 4 patients died: 2 from the LDLT and 2 from the DDLT (3 who met the Milan criteria and 1 who met the UCSF criteria), and none experienced HCC recurrence. Two patients from the positive FDG-PET CT group died, 3 and 87 days after LT, owing to primary graft nonfunction and infection. The remaining 29 patients (70.7%) had negative FDG-PET CT results. Of these, 21 patients met the Milan criteria, 5 patients met the UCSF criteria, and 3 patients exceeded the UCSF criteria. Of the 29 patients, the explant showed well-differentiated HCC in 8 patients, moderately differentiated HCC in 12 patients, mixed HCC and cholangiocarcinoma in 1 patient, and complete necrosis due to successful locoregional therapy before LT in 8 patients. Two patients from negative FDG-PET CT group died of infection and hepatic artery thrombosis, respectively.
The sensitivity, specificity, positive predictive value, and negative predictive value of FDG-PET CT for detecting HCC before LT were found to be 27.5%, 100%, 100%, and 3.3%, respectively (Table 1). The diagnostic accuracy of FDG-PET CT was found to increase as tumor size increases: to 16% for patients with HCC that met the Milan criteria, 40% for patients with HCC that met the UCSF criteria and 62% for patients with HCC that exceeded the UCSF criteria (Table 2). The median 3-year survival rates for patients with HCC were 81.8% and 93.0% for the positive and negative FDG-PET CT groups.
Several investigators have reported controversial conclusions and inadequate sensitivity of PET for detecting various malignancies11-12,14,19-21. In evaluating the role of FDG-PET CT for detecting HCC in pre-LT patients, we found our results to be in agreement with other reports that evaluated the role of PET scans in patients with HCC. We have shown that the FDG-PET CT scan has low use in patients whose HCC meets the Milan criteria and should not be routinely used as part of the pre-LT work-up. However, FDG-PET CT may have a role in detecting tumors in patients whose HCC exceeds the UCSF criteria.
In a recent report, Kornberg and associates retrospectively analyzed the role of preoperative PET using (18)F-FDG for predicting microvascular tumor invasion and posttransplant tumor recurrence in LT candidates with HCC. In their study, patients whose preoperative PET results were negative demonstrated significantly better 3-year recurrence-free survival (93%) than patients with positive PET results before LT (35%; P < .001). The HCC recurrence rate was 50% in the positive-result group but only 3.8% in the negative-result group (P < .001). Further, positive PET CT status was identified as an independent predictor of microvascular tumor invasion.22 However, the authors concluded that PET scanning was more useful in patients whose HCC exceeded the Milan criteria. In our study, positive FDG-PET CT results were not associated with microvascular tumor invasion, poor tumor differentiation or HCC recurrence after LT.
In another report, Yang and colleagues retrospectively reviewed 38 HCC patients who underwent LT and whole-body PET imaging at their center. (18)F-FDG uptake was assessed in the liver, and its prognostic significance was investigated. In their series, 13 patients had a positive PET scan result. They analyzed the association between tumor factors and positive PET scan results in the liver; the preoperative alpha-fetoprotein level and vascular invasion were found to be significantly associated with positive PET scan results (P = .003 and P < .001). In agreement with this study, the association between histologic grade and positive PET scan results was not significant. The 2-year recurrence-free survival rate of patients with negative PET scans was significantly higher than for patients with positive PET scans (85.1% vs. 46.1%; P < .001). Of 6 patients with positive PET scans who met the Milan criteria, 4 patients (66.7%) had recurrence; in contrast, all 20 patients with negative PET scans who met the Milan criteria were recurrence-free. Thus, the authors concluded that PET imaging could be a good preoperative tool for estimating the post-LT risk of tumor recurrence.23
Böhm and associates studied the efficacy of PET scanning for evaluating patients before liver resection to treat primary and secondary liver lesions. They concluded that the main value of PET scanning is for the detection of extrahepatic malignancy.24 Liangpunsakul and colleagues hypothesized that PET scanning might be useful for the detection of occult HCC in patients with cirrhosis due to hepatitis C virus. Positron emission tomography scanning was performed in 8 such patients who were on the LT waiting list and displayed persistently elevated alpha-fetoprotein levels (> 100 ng/mL) but no detectable lesions on an abdominal CT scan. The results of PET detection of occult HCC were compared with those obtained using lipiodol-enhanced CT scanning and histologic examination of the live explant. Explant histology and prolonged clinical follow-up showed that 2 study subjects had conclusive evidence of HCC; the remaining 6 patients had no evidence of malignancy. In this study, PET imaging did not reveal abnormal lesions in any subject; lipiodol-enhanced CT scans revealed abnormal lipiodol retention in the 2 subjects with HCC.25
Our series revealed that FDG-PET CT scanning has a sensitivity of 27.5%, a specificity of 100%, a positive predictive value of 100% and a negative predictive value of 3.3% for detecting HCC prior to LT. Thus, FDG-PET CT has low use in patients whose HCC meets the Milan criteria and should not be routinely used as part of an LT work-up. However, FDG-PET CT may have use in patients whose HCC exceeds the UCSF criteria. In addition, it is not known whether using 18F-fluorocholine PET CT in patients with negative FDG-PET CT scans might add diagnostic value, so this protocol needs to be studied prospectively.
Volume : 15
Issue : 1
Pages : 37 - 41
DOI : 10.6002/ect.mesot2016.O21
From the 1Department of Liver & Small Bowel Transplantation &
Hepatobiliary-Pancreatic Surgery, King Faisal Specialist Hospital & Research
Centre, Riyadh, Saudi Arabia; 2College of Medicine, Alfaisal
University, Riyadh, Saudi Arabia; and 3Division of Gastroenterology,
King Saud University, Riyadh, Saudi Arabia
Acknowledgements: The authors declare that they have no sources of funding for this study, and they have no conflicts of interest to declare.
Corresponding author: Hussien Elsiesy, Department of Liver & Small Bowel Transplantation & Hepatobiliary-Pancreatic Surgery, King Faisal Specialist Hospital & Research Centre, Altakhasusi Street, Riyadh, 11211, Saudi Arabia
Phone: +966 11 442 4818
Table 1. Accuracy of FDG-PET CT Scans* for Detecting HCC in Patients Before Liver Transplant (n = 40)
Table 2. FDG-PET CT Results of Patients Prior to Liver Transplant (n = 40) in Relation to Tumor Size