Original Clinical ScienceçLiver
Clinical Impact of 18F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography in Living Donor Liver Transplantation for Advanced Hepatocellular Carcinoma Seung Duk Lee, MD,1 Seong Hoon Kim, MD, PhD,1 Seok-Ki Kim, MD, PhD,2 Young-Kyu Kim, MD,3 and Sang-Jae Park, MD, PhD1
Background. The relevant number of patients with hepatocellular carcinoma (HCC) beyond the Milan criteria have undergone living donor liver transplantation (LDLT). However, the prognostic factors for these patients with advanced HCC remain unclear. Methods. From March 2005 to May 2013, 280 patients with HCC underwent LDLTat the National Cancer Center. Of these, patients with HCC beyond the Milan criteria were retrospectively enrolled. We analyzed the prognostic significance of 18 F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG-PET/CT) for selecting appropriate candidates. Results. Of the total 280 patients, 147 (52.5%) were confirmed to have HCC beyond the Milan criteria on the basis of pathological reports. The patients who met and exceeded the Milan criteria had 5-year overall survival (OS) rates of 87.2% and 64.6%, respectively (P < 0.001). Multivariable analysis for OS and disease-free survival (DFS) in patients with HCC beyond the Milan criteria revealed PET/CT positivity (hazards ratio [HR], 2.714; P = 0.013 for OS; HR, 3.803; P < 0.001 for DFS), total tumor size over 10 cm (HR, 2.333; P = 0.035 for OS; HR, 3.334, P = 0.001 for DFS), and microvascular invasion (HR, 2.917; P = 0.025 for DFS) to be significant prognostic factors. In particular, patients with HCC beyond the Milan criteria with a PET/CT-negative status and total tumor size less than 10 cm showed similar OS and DFS in comparison with those with HCC within the Milan criteria. Conclusions. A PET/CT status in LDLT is a useful marker for predicting survival of patients with advanced HCC.
(Transplantation 2015;99: 2142–2149)
L
iving donor liver transplantation (LDLT) has been accepted as an effective therapy for hepatocellular carcinoma (HCC) when there is a critical shortage of deceased organs.1 In LDLT, the issue of selection criteria including
Received 4 September 2014. Revision requested 6 January 2015. Accepted 9 January 2015. 1
Center for Liver Cancer, National Cancer Center, Goyang-si, Gyeonggi-do, Republic of Korea. 2 Department of Nuclear Medicine, National Cancer Center, Goyang-si, Gyeonggi-do, Republic of Korea. 3
Department of Surgery, College of Medicine, Jeju National University, Jeju-do, Republic of Korea. This study was supported by National Cancer Center, Republic of Korea (grants 1410910-2 and 1310170-3). The authors declare no conflicts of interest. S.D.L. collected and analyzed the data and wrote the article. S.H.K. designed study and performed the study. S.-K.K. Performed the study and analyzed data. Y.-K.K. performed the study and collected data. S.-J.P. performed the study and contributed important reagents. Correspondence: Seong Hoon Kim, MD, PhD, Center for Liver Cancer, Research Institute and Hospital, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si Gyeonggi-do, 410-769, Republic of Korea. ([email protected]). Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0041-1337/15/9910-2142 DOI: 10.1097/TP.0000000000000719
2142
www.transplantjournal.com
the Milan (a solitary tumor no more than 5 cm in diameter, or 2 or 3 tumors no more than 3 cm in diameter) and the University of California at San Francisco (UCSF, a solitary tumor no more than 6.5 cm in diameter or 2 or 3 tumors with the largest diameter being no more than 4.5 cm and the sum of the diameters being no more than 8 cm) criteria, has been continued.2,3 These criteria using tumor number and size are very useful and have been widely used; however, they may be too restrictive to correctly reflect tumor characteristics. The definition of advanced HCC involves preoperative tumor number and size on computed tomography (CT) or MRI. However, preoperative imaging techniques had limitation with regard to accurate tumor characterization because of the differences in pathological data and complicated readings by preoperative local therapy and transarterial chemoembolization (TACE).4 Therefore, large numbers of patients with advanced HCC with good tumor biology are being excluded from liver transplantation (LT) and miss the chance of curative modality.5,6 Expanded criteria, such as 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/ CT), α-fetoprotein (AFP), protein induced by vitamin K absence or antagonist-II (PIVKA-II), and tumor morphology have recently been proposed.6-10 There is now evidence that the biological features of a tumor, rather than its macromorphology, are associated with patient prognosis and outcomes in advanced HCC after LT.11,12 Pathological factors, including tumor differentiation grade and microvascular invasion, are well-known independent prognostic markers Transplantation
■
October 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
■
Volume 99
■
Number 10
Lee et al
© 2015 Wolters Kluwer
used after LT; however, these could not be predicted using only tumor morphology before LT.13 One study performed preoperative 18F-FDG PET before LT and concluded that 18 F-FDG uptake of HCC predicted microvascular tumor invasion.14 The PET/CT before surgery is also a good diagnostic tool for detecting metastatic lesions.15 There are few prognostic studies using PET/CT for advanced HCC and LDLT. In this study, we performed a retrospective analysis to evaluate the prognostic significance of PET/CT and tumor morphology findings in a relatively large number of LDLT patients with pathologically confirmed advanced HCC. PATIENTS AND METHODS Patients
Patients who underwent preoperative PET/CT and LDLT for HCC between March 2005 and May 2013 were enrolled from our prospective database at the National Cancer Center, Republic of Korea. All the patients were confirmed to have HCC using pathological reports. Preoperative clinical staging for patients with HCC included ultrasonography, multidetector CT, and dual contrast-enhanced MRI without protocol tumor biopsy. Patients determined to have HCC with the abovementioned imaging modalities underwent whole-body 18F-FDG PET/CT within 1 month before surgery. The medical records of the patients were reviewed for clinicopathological data, including sex, age, viral marker, serum AFP, C-reactive protein, model for end-stage liver disease score, PET/CT positivity, tumor maximum standardized uptake value (SUV), preoperative therapy, and pathological reports. The pathological data included tumor differentiation using Edmondson and Steiner grade; microvascular, major vessel, ductal, or serosal invasion; capsule formation; intrahepatic metastasis; cirrhosis; and dysplastic nodules. The Milan and UCSF criteria in this study were determined by information on pathological reports. The mean followup period was 37.3 ± 24.1 months. The study protocol was approved by the institutional review board of our hospital. Patient Selection and Follow-up
Our policy of selecting patients with HCC for LDLT was based on preoperative imaging studies (e.g., CT, MRI, or PET/CT) that met the Milan criteria. However, if patients were determined to be beyond the Milan criteria without major vascular invasion and extrahepatic metastasis and strongly desired LDLT, we performed LDLT but provided detailed information regarding the possibility of tumor
FIGURE 1.
2143
recurrence. All studies, including imaging and laboratory tests, were usually performed within 1 month before LDLT. No downstaging or bridging therapy was used for LDLT at our hospital. We mostly used the donor right lobe without the middle hepatic vein for LDLT, and reconstructed middle hepatic vein with an artificial vascular graft. The detailed surgical technique for LDLT has been described previously.16 Recently, the morbidity rate of donor hepatectomy was 3.0% without the major complications.17 Immunosuppressive therapy after LDLT included a triple regimen with tacrolimus, mycophenolate mofetil, and corticosteroid. The latter was tapered to discontinuation by 3 months after LDLT. The prophylaxis for hepatitis B virus recurrence consisted of combination drugs with antiviral agents and hepatitis B immunoglobulin. The treatment of hepatitis C virus recurrence was initiated with peginterferon α and ribavirin after pathological and laboratory diagnosis. Patients with HCC were followed up after LDLT every 3 months for the first 2 years and every 6 months for the next 3 years. Imaging studies, including abdomen and chest CT and bone scans, were performed every 3 or 6 months with AFP and PIVKA-II level assessments. When the tumor recurrence was detected by imaging and laboratory tests, additional whole-body PET/CT was performed to evaluate distant metastases. We performed resections for 1 or 2 nodules in the liver, lung, bone, or brain without multiple metastases. However, for multiple recurrence, we treated tumors with a multidisciplinary approach, including TACE, radiofrequency ablation, radiation therapy, or chemotherapy (e.g., sorafenib). 18
F-FDG PET/CT
We previously described a detailed 18F-FDG PET/CT protocol.8 In brief, all PET/CT images were interpreted by experienced nuclear medicine physicians at our hospital, and SUV was calculated as (decay-corrected activity kBq/mL of tissue volume)/(injected FDG activity kBq/body mass g). Standardized uptake values of the lesions were checked by manually placing a circular region of interest (ROI) at the site of the maximum FDG uptake in the smoothed PET images. All tumors and normal liver regions were defined by careful correlation with previous CT or MRI, and ROIs were drawn to encircle the highest activity of each tumor. The PET/CT positivity was determined by experienced nuclear physicians who examined whether the FDG uptake in tumor was significantly higher than that in the surrounding noncancerous hepatic tissue (Figure 1). We used maximum SUV (SUVmax)
18
F-FDG-PET/CT and MRI images showing advanced HCC before liver transplantation (tumor size, 11.5 cm; SUVmax, 6.86).
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
2144
Transplantation
■
October 2015
■
Volume 99
■
Number 10
within ROI in this study. Mean SUVmax for tumors defined as PET/CT positivity and negativity was 4.46 and 3.08, respectively (P < 0.001). The mean period between PET/CT and LDLT was 14.8 ± 23.1 days. Statistical Analysis 2
Clinicopathological data were analyzed with χ tests or Fisher exact tests for categorical variables and Student t tests for continuous variables. Overall survival (OS) and diseasefree survival (DFS) were analyzed using the Kaplan-Meier method and log-rank test. The prognostic factors were analyzed using a Cox proportional hazard model. Variables found to be significant on univariate analysis (P < 0.1) were considered in a multivariable model. A P value less than 0.05 was considered statistically significant, and all analyses were performed using SAS version 9.1.3 for Windows (SAS institute, Cary, NC). RESULTS Clinicopathological Characteristics
A total of 291 patients with HCC underwent LDLT during the study period. Eleven patients were excluded from the analysis because their data of PET/CT were not available. Among them, 147 (52.5%) patients belonged and exceeded the Milan criteria. Table 1 compares the clinicopathological data in patients within and beyond the Milan criteria. Male sex, C-reactive protein, tumor SUVmax, AFP (>400 ng/mL), PET/CT positivity, differentiation (grades III-IV), microvascular invasion, serosal invasion, and intrahepatic metastasis were observed at significantly higher levels in patients beyond the Milan criteria than in those within the Milan criteria. Mean levels of C-reactive protein (mg/dL) and tumor SUVmax in patients within and beyond the Milan criteria were 0.61/3.23 and 1.17/3.78, respectively. The PET/CT positivity was noted in 45.6% of patients beyond the Milan criteria in comparison with 17.3% in patients within the Milan criteria. The type of viral hepatitis, presence of capsule formation, ductal invasion, cirrhosis, and dysplastic nodules were not significantly different between the 2 groups. Survivals and Tumor Variables by PET/CT Positivity in Advanced HCC
The 5-year OS rates between patients within and beyond the Milan criteria were 87.2% and 64.6%, respectively, and DFS rates were 89.8% and 56.7%, respectively (Figure S1, SDC, http://links.lww.com/TP/B145). Patients beyond the Milan criteria who also had a PET/CT-positive status showed significantly worse survivals than those who had a PET/CTnegative status (5-year OS, 74.6% vs 51.4%, P < 0.001; 5-year DFS, 73.3% vs 37.5%, P < 0.001; Figure 2). However, in patients within the Milan criteria, only the DFS rate was significantly different between groups by the PET/CT status (5-year DFS, 92.3% vs 76.3%, P = 0.031). Similarly, survival analyses based on the UCSF criteria and PET/CT findings revealed that OS and DFS rates in patients beyond the UCSF criteria with a PET/CT-positive status were significantly worse than those in patients with a PET/CTnegative status (5-year OS, 74.6% vs 44.5%, P < 0.001; 5-year DFS, 72.8% vs 30.7%, P < 0.001; Figure 3). In patients within the USCF criteria, there were no differences in OS or DFS rates according to PET/CT positivity (OS: mean, 91.9 vs 90.4 months, P = 0.844; DFS: mean, 96.1 vs 87.0 months,
www.transplantjournal.com
TABLE 1.
Clinicopathologic characteristics of patients according to Milan criteria Within Milan Beyond Milan (n = 133) (n = 147)
Variables
Sex, n (%)
Male Female
Age: mean (SD), y MELD score, mean (SD) C-reactive protein: mean (SD), mg/dL Tumor maximum SUV, mean (SD) AFP, n (%)
≤400 ng/mL >400 ng/mL PET/CT Negative positive Pretransplant No therapy therapy, n (%) Surgery only TACE only RFA only Combination Viral hepatitis HBV HCV NBNC HBV + HCV Differentiationa, n (%) I-II III-IV Microvascular Absent invasion, n (%) Present Capsule No complete formation, n (%) Complete Ductal invasion, n (%) Absent Present Serosal invasion, n (%) Absent Present Intrahepatic Absent metastasis, n (%) Present Cirrhosis, n (%) Absent Present Dysplastic Absent nodule, n (%) Present
P
103 (77.4) 30 (22.6) 53.8 (7.0) 14.4 (7.7)
132 (89.8) 15 (10.2) 55.0 (7.5) 12.9 (6.8)
0.005 0.162 0.081
0.61 (1.11)
1.17 (2.43)
0.014
3.23 (1.33)
3.78 (1.33)
0.001
124 (93.2) 9 (6.8) 110 (82.7) 23 (17.3) 38 (28.6) 7 (5.3) 50 (37.6) 7 (5.3) 31 (23.3) 116 (87.2) 7 (5.3) 9 (6.8) 1 (0.8) 88 (66.2) 45 (33.8) 104 (78.2) 29 (21.8) 108 (81.2) 25 (18.8) 129 (97.0) 4 (3.0) 113 (85.0) 20 (15.0) 116 (87.2) 17 (12.8) 9 (6.8) 124 (93.2) 101 (75.9) 32 (24.1)
115 (78.2) 32 (21.8) 80 (54.4) 67 (45.6) 30 (20.4) 5 (3.4) 66 (44.9) 1 (0.7) 45 (30.6) 129 (87.8) 10 (6.8) 5 (3.4) 3 (2.0) 69 (46.9) 78 (53.1) 70 (47.6) 77 (52.4) 120 (81.6) 27 (18.4) 141 (95.9) 6 (4.1) 105 (71.4) 42 (28.6) 68 (46.3) 79 (53.7) 12 (8.2) 135 (91.8) 100 (68.0) 47 (32.0)
10 cm); and microvascular, major vessel, serosal invasion.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Lee et al
© 2015 Wolters Kluwer
2145
FIGURE 2. Overall and disease-free survival rates according to the Milan criteria and PET/CT positivity.
Prognostic Factors in Patients Beyond the Milan Criteria
The results of univariate and multivariable analyses for OS and DFS in patients beyond the Milan criteria are summarized in Table 3. In univariate analysis, the AFP level (>400 ng/mL), PET/CT positivity, maximum tumor size (>5 cm), total tumor size (>10 cm), microvascular invasion, major vessel invasion, serosal invasion, intrahepatic metastasis,
and dysplastic nodules were significant prognostic factors for both OS and DFS. Tumor number (>3 cm), differentiation, and ductal invasion were only significant prognostic factors for DFS. In multivariable analysis, PET/CT positivity and total tumor size (>10 cm) were statistically significant prognostic factors for both OS and DFS (OS: PET/CT positivity, hazards ratio [HR], 2.714, P = 0.013; total tumor size, HR,
FIGURE 3. Overall and disease-free survival rates according to the UCSF criteria and PET/CT positivity.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
2146
Transplantation
■
October 2015
■
Volume 99
■
Number 10
TABLE 2.
Association between PET CT findings and tumor characteristics beyond the Milan criteria Variables
≤400 ng/mL >400 ng/mL Tumor ≤3 number, n (%) >3 Maximum ≤5 cm tumor size >5 cm Total tumor size ≤10 cm >10 cm Differentiationa I-II III-IV Microvascular Absent invasion, n (%) Present Capsule No complete formation, n (%) Complete Major vessel Absent invasion, n (%) Present Ductal Absent invasion, n (%) Present Serosal Absent invasion, n (%) Present Intrahepatic Absent metastasis, n (%) Present AFP, n (%)
a
PET/CT PET/CT negative, n = 80 positive, n = 67
72 (90.0) 8 (10.0) 17 (21.2) 63 (78.8) 66 (82.5) 14 (17.5) 55 (68.8) 25 (31.2) 43 (53.8) 37 (46.2) 48 (60.0) 32 (40.0) 60 (75.0) 20 (25.0) 78 (97.5) 2 (2.5) 78 (97.5) 2 (2.5) 70 (87.5) 10 (12.5) 39 (48.8) 41 (51.2)
43 (64.2) 24 (35.8) 30 (44.8) 37 (55.2) 34 (50.7) 33 (49.3) 32 (47.8) 35 (52.2) 26 (38.8) 41 (61.2) 22 (32.8) 45 (67.2) 58 (86.6) 9 (13.4) 49 (73.1) 18 (26.9) 63 (94.0) 4 (6.0) 35 (52.2) 32 (47.8) 29 (43.3) 38 (56.7)
P
Clinical Impact of 18F-Fluorodeoxyglucose Positron Emission Tomography/Computed Tomography in Living Donor Liver Transplantation for Advanced Hepatocellular Carcinoma Seung Duk Lee, MD,1 Seong Hoon Kim, MD, PhD,1 Seok-Ki Kim, MD, PhD,2 Young-Kyu Kim, MD,3 and Sang-Jae Park, MD, PhD1
Background. The relevant number of patients with hepatocellular carcinoma (HCC) beyond the Milan criteria have undergone living donor liver transplantation (LDLT). However, the prognostic factors for these patients with advanced HCC remain unclear. Methods. From March 2005 to May 2013, 280 patients with HCC underwent LDLTat the National Cancer Center. Of these, patients with HCC beyond the Milan criteria were retrospectively enrolled. We analyzed the prognostic significance of 18 F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG-PET/CT) for selecting appropriate candidates. Results. Of the total 280 patients, 147 (52.5%) were confirmed to have HCC beyond the Milan criteria on the basis of pathological reports. The patients who met and exceeded the Milan criteria had 5-year overall survival (OS) rates of 87.2% and 64.6%, respectively (P < 0.001). Multivariable analysis for OS and disease-free survival (DFS) in patients with HCC beyond the Milan criteria revealed PET/CT positivity (hazards ratio [HR], 2.714; P = 0.013 for OS; HR, 3.803; P < 0.001 for DFS), total tumor size over 10 cm (HR, 2.333; P = 0.035 for OS; HR, 3.334, P = 0.001 for DFS), and microvascular invasion (HR, 2.917; P = 0.025 for DFS) to be significant prognostic factors. In particular, patients with HCC beyond the Milan criteria with a PET/CT-negative status and total tumor size less than 10 cm showed similar OS and DFS in comparison with those with HCC within the Milan criteria. Conclusions. A PET/CT status in LDLT is a useful marker for predicting survival of patients with advanced HCC.
(Transplantation 2015;99: 2142–2149)
L
iving donor liver transplantation (LDLT) has been accepted as an effective therapy for hepatocellular carcinoma (HCC) when there is a critical shortage of deceased organs.1 In LDLT, the issue of selection criteria including
Received 4 September 2014. Revision requested 6 January 2015. Accepted 9 January 2015. 1
Center for Liver Cancer, National Cancer Center, Goyang-si, Gyeonggi-do, Republic of Korea. 2 Department of Nuclear Medicine, National Cancer Center, Goyang-si, Gyeonggi-do, Republic of Korea. 3
Department of Surgery, College of Medicine, Jeju National University, Jeju-do, Republic of Korea. This study was supported by National Cancer Center, Republic of Korea (grants 1410910-2 and 1310170-3). The authors declare no conflicts of interest. S.D.L. collected and analyzed the data and wrote the article. S.H.K. designed study and performed the study. S.-K.K. Performed the study and analyzed data. Y.-K.K. performed the study and collected data. S.-J.P. performed the study and contributed important reagents. Correspondence: Seong Hoon Kim, MD, PhD, Center for Liver Cancer, Research Institute and Hospital, National Cancer Center, 323 Ilsan-ro, Ilsandong-gu, Goyang-si Gyeonggi-do, 410-769, Republic of Korea. ([email protected]). Supplemental digital content (SDC) is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s Web site (www.transplantjournal.com). Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. ISSN: 0041-1337/15/9910-2142 DOI: 10.1097/TP.0000000000000719
2142
www.transplantjournal.com
the Milan (a solitary tumor no more than 5 cm in diameter, or 2 or 3 tumors no more than 3 cm in diameter) and the University of California at San Francisco (UCSF, a solitary tumor no more than 6.5 cm in diameter or 2 or 3 tumors with the largest diameter being no more than 4.5 cm and the sum of the diameters being no more than 8 cm) criteria, has been continued.2,3 These criteria using tumor number and size are very useful and have been widely used; however, they may be too restrictive to correctly reflect tumor characteristics. The definition of advanced HCC involves preoperative tumor number and size on computed tomography (CT) or MRI. However, preoperative imaging techniques had limitation with regard to accurate tumor characterization because of the differences in pathological data and complicated readings by preoperative local therapy and transarterial chemoembolization (TACE).4 Therefore, large numbers of patients with advanced HCC with good tumor biology are being excluded from liver transplantation (LT) and miss the chance of curative modality.5,6 Expanded criteria, such as 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/ CT), α-fetoprotein (AFP), protein induced by vitamin K absence or antagonist-II (PIVKA-II), and tumor morphology have recently been proposed.6-10 There is now evidence that the biological features of a tumor, rather than its macromorphology, are associated with patient prognosis and outcomes in advanced HCC after LT.11,12 Pathological factors, including tumor differentiation grade and microvascular invasion, are well-known independent prognostic markers Transplantation
■
October 2015
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
■
Volume 99
■
Number 10
Lee et al
© 2015 Wolters Kluwer
used after LT; however, these could not be predicted using only tumor morphology before LT.13 One study performed preoperative 18F-FDG PET before LT and concluded that 18 F-FDG uptake of HCC predicted microvascular tumor invasion.14 The PET/CT before surgery is also a good diagnostic tool for detecting metastatic lesions.15 There are few prognostic studies using PET/CT for advanced HCC and LDLT. In this study, we performed a retrospective analysis to evaluate the prognostic significance of PET/CT and tumor morphology findings in a relatively large number of LDLT patients with pathologically confirmed advanced HCC. PATIENTS AND METHODS Patients
Patients who underwent preoperative PET/CT and LDLT for HCC between March 2005 and May 2013 were enrolled from our prospective database at the National Cancer Center, Republic of Korea. All the patients were confirmed to have HCC using pathological reports. Preoperative clinical staging for patients with HCC included ultrasonography, multidetector CT, and dual contrast-enhanced MRI without protocol tumor biopsy. Patients determined to have HCC with the abovementioned imaging modalities underwent whole-body 18F-FDG PET/CT within 1 month before surgery. The medical records of the patients were reviewed for clinicopathological data, including sex, age, viral marker, serum AFP, C-reactive protein, model for end-stage liver disease score, PET/CT positivity, tumor maximum standardized uptake value (SUV), preoperative therapy, and pathological reports. The pathological data included tumor differentiation using Edmondson and Steiner grade; microvascular, major vessel, ductal, or serosal invasion; capsule formation; intrahepatic metastasis; cirrhosis; and dysplastic nodules. The Milan and UCSF criteria in this study were determined by information on pathological reports. The mean followup period was 37.3 ± 24.1 months. The study protocol was approved by the institutional review board of our hospital. Patient Selection and Follow-up
Our policy of selecting patients with HCC for LDLT was based on preoperative imaging studies (e.g., CT, MRI, or PET/CT) that met the Milan criteria. However, if patients were determined to be beyond the Milan criteria without major vascular invasion and extrahepatic metastasis and strongly desired LDLT, we performed LDLT but provided detailed information regarding the possibility of tumor
FIGURE 1.
2143
recurrence. All studies, including imaging and laboratory tests, were usually performed within 1 month before LDLT. No downstaging or bridging therapy was used for LDLT at our hospital. We mostly used the donor right lobe without the middle hepatic vein for LDLT, and reconstructed middle hepatic vein with an artificial vascular graft. The detailed surgical technique for LDLT has been described previously.16 Recently, the morbidity rate of donor hepatectomy was 3.0% without the major complications.17 Immunosuppressive therapy after LDLT included a triple regimen with tacrolimus, mycophenolate mofetil, and corticosteroid. The latter was tapered to discontinuation by 3 months after LDLT. The prophylaxis for hepatitis B virus recurrence consisted of combination drugs with antiviral agents and hepatitis B immunoglobulin. The treatment of hepatitis C virus recurrence was initiated with peginterferon α and ribavirin after pathological and laboratory diagnosis. Patients with HCC were followed up after LDLT every 3 months for the first 2 years and every 6 months for the next 3 years. Imaging studies, including abdomen and chest CT and bone scans, were performed every 3 or 6 months with AFP and PIVKA-II level assessments. When the tumor recurrence was detected by imaging and laboratory tests, additional whole-body PET/CT was performed to evaluate distant metastases. We performed resections for 1 or 2 nodules in the liver, lung, bone, or brain without multiple metastases. However, for multiple recurrence, we treated tumors with a multidisciplinary approach, including TACE, radiofrequency ablation, radiation therapy, or chemotherapy (e.g., sorafenib). 18
F-FDG PET/CT
We previously described a detailed 18F-FDG PET/CT protocol.8 In brief, all PET/CT images were interpreted by experienced nuclear medicine physicians at our hospital, and SUV was calculated as (decay-corrected activity kBq/mL of tissue volume)/(injected FDG activity kBq/body mass g). Standardized uptake values of the lesions were checked by manually placing a circular region of interest (ROI) at the site of the maximum FDG uptake in the smoothed PET images. All tumors and normal liver regions were defined by careful correlation with previous CT or MRI, and ROIs were drawn to encircle the highest activity of each tumor. The PET/CT positivity was determined by experienced nuclear physicians who examined whether the FDG uptake in tumor was significantly higher than that in the surrounding noncancerous hepatic tissue (Figure 1). We used maximum SUV (SUVmax)
18
F-FDG-PET/CT and MRI images showing advanced HCC before liver transplantation (tumor size, 11.5 cm; SUVmax, 6.86).
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
2144
Transplantation
■
October 2015
■
Volume 99
■
Number 10
within ROI in this study. Mean SUVmax for tumors defined as PET/CT positivity and negativity was 4.46 and 3.08, respectively (P < 0.001). The mean period between PET/CT and LDLT was 14.8 ± 23.1 days. Statistical Analysis 2
Clinicopathological data were analyzed with χ tests or Fisher exact tests for categorical variables and Student t tests for continuous variables. Overall survival (OS) and diseasefree survival (DFS) were analyzed using the Kaplan-Meier method and log-rank test. The prognostic factors were analyzed using a Cox proportional hazard model. Variables found to be significant on univariate analysis (P < 0.1) were considered in a multivariable model. A P value less than 0.05 was considered statistically significant, and all analyses were performed using SAS version 9.1.3 for Windows (SAS institute, Cary, NC). RESULTS Clinicopathological Characteristics
A total of 291 patients with HCC underwent LDLT during the study period. Eleven patients were excluded from the analysis because their data of PET/CT were not available. Among them, 147 (52.5%) patients belonged and exceeded the Milan criteria. Table 1 compares the clinicopathological data in patients within and beyond the Milan criteria. Male sex, C-reactive protein, tumor SUVmax, AFP (>400 ng/mL), PET/CT positivity, differentiation (grades III-IV), microvascular invasion, serosal invasion, and intrahepatic metastasis were observed at significantly higher levels in patients beyond the Milan criteria than in those within the Milan criteria. Mean levels of C-reactive protein (mg/dL) and tumor SUVmax in patients within and beyond the Milan criteria were 0.61/3.23 and 1.17/3.78, respectively. The PET/CT positivity was noted in 45.6% of patients beyond the Milan criteria in comparison with 17.3% in patients within the Milan criteria. The type of viral hepatitis, presence of capsule formation, ductal invasion, cirrhosis, and dysplastic nodules were not significantly different between the 2 groups. Survivals and Tumor Variables by PET/CT Positivity in Advanced HCC
The 5-year OS rates between patients within and beyond the Milan criteria were 87.2% and 64.6%, respectively, and DFS rates were 89.8% and 56.7%, respectively (Figure S1, SDC, http://links.lww.com/TP/B145). Patients beyond the Milan criteria who also had a PET/CT-positive status showed significantly worse survivals than those who had a PET/CTnegative status (5-year OS, 74.6% vs 51.4%, P < 0.001; 5-year DFS, 73.3% vs 37.5%, P < 0.001; Figure 2). However, in patients within the Milan criteria, only the DFS rate was significantly different between groups by the PET/CT status (5-year DFS, 92.3% vs 76.3%, P = 0.031). Similarly, survival analyses based on the UCSF criteria and PET/CT findings revealed that OS and DFS rates in patients beyond the UCSF criteria with a PET/CT-positive status were significantly worse than those in patients with a PET/CTnegative status (5-year OS, 74.6% vs 44.5%, P < 0.001; 5-year DFS, 72.8% vs 30.7%, P < 0.001; Figure 3). In patients within the USCF criteria, there were no differences in OS or DFS rates according to PET/CT positivity (OS: mean, 91.9 vs 90.4 months, P = 0.844; DFS: mean, 96.1 vs 87.0 months,
www.transplantjournal.com
TABLE 1.
Clinicopathologic characteristics of patients according to Milan criteria Within Milan Beyond Milan (n = 133) (n = 147)
Variables
Sex, n (%)
Male Female
Age: mean (SD), y MELD score, mean (SD) C-reactive protein: mean (SD), mg/dL Tumor maximum SUV, mean (SD) AFP, n (%)
≤400 ng/mL >400 ng/mL PET/CT Negative positive Pretransplant No therapy therapy, n (%) Surgery only TACE only RFA only Combination Viral hepatitis HBV HCV NBNC HBV + HCV Differentiationa, n (%) I-II III-IV Microvascular Absent invasion, n (%) Present Capsule No complete formation, n (%) Complete Ductal invasion, n (%) Absent Present Serosal invasion, n (%) Absent Present Intrahepatic Absent metastasis, n (%) Present Cirrhosis, n (%) Absent Present Dysplastic Absent nodule, n (%) Present
P
103 (77.4) 30 (22.6) 53.8 (7.0) 14.4 (7.7)
132 (89.8) 15 (10.2) 55.0 (7.5) 12.9 (6.8)
0.005 0.162 0.081
0.61 (1.11)
1.17 (2.43)
0.014
3.23 (1.33)
3.78 (1.33)
0.001
124 (93.2) 9 (6.8) 110 (82.7) 23 (17.3) 38 (28.6) 7 (5.3) 50 (37.6) 7 (5.3) 31 (23.3) 116 (87.2) 7 (5.3) 9 (6.8) 1 (0.8) 88 (66.2) 45 (33.8) 104 (78.2) 29 (21.8) 108 (81.2) 25 (18.8) 129 (97.0) 4 (3.0) 113 (85.0) 20 (15.0) 116 (87.2) 17 (12.8) 9 (6.8) 124 (93.2) 101 (75.9) 32 (24.1)
115 (78.2) 32 (21.8) 80 (54.4) 67 (45.6) 30 (20.4) 5 (3.4) 66 (44.9) 1 (0.7) 45 (30.6) 129 (87.8) 10 (6.8) 5 (3.4) 3 (2.0) 69 (46.9) 78 (53.1) 70 (47.6) 77 (52.4) 120 (81.6) 27 (18.4) 141 (95.9) 6 (4.1) 105 (71.4) 42 (28.6) 68 (46.3) 79 (53.7) 12 (8.2) 135 (91.8) 100 (68.0) 47 (32.0)
10 cm); and microvascular, major vessel, serosal invasion.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
Lee et al
© 2015 Wolters Kluwer
2145
FIGURE 2. Overall and disease-free survival rates according to the Milan criteria and PET/CT positivity.
Prognostic Factors in Patients Beyond the Milan Criteria
The results of univariate and multivariable analyses for OS and DFS in patients beyond the Milan criteria are summarized in Table 3. In univariate analysis, the AFP level (>400 ng/mL), PET/CT positivity, maximum tumor size (>5 cm), total tumor size (>10 cm), microvascular invasion, major vessel invasion, serosal invasion, intrahepatic metastasis,
and dysplastic nodules were significant prognostic factors for both OS and DFS. Tumor number (>3 cm), differentiation, and ductal invasion were only significant prognostic factors for DFS. In multivariable analysis, PET/CT positivity and total tumor size (>10 cm) were statistically significant prognostic factors for both OS and DFS (OS: PET/CT positivity, hazards ratio [HR], 2.714, P = 0.013; total tumor size, HR,
FIGURE 3. Overall and disease-free survival rates according to the UCSF criteria and PET/CT positivity.
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
2146
Transplantation
■
October 2015
■
Volume 99
■
Number 10
TABLE 2.
Association between PET CT findings and tumor characteristics beyond the Milan criteria Variables
≤400 ng/mL >400 ng/mL Tumor ≤3 number, n (%) >3 Maximum ≤5 cm tumor size >5 cm Total tumor size ≤10 cm >10 cm Differentiationa I-II III-IV Microvascular Absent invasion, n (%) Present Capsule No complete formation, n (%) Complete Major vessel Absent invasion, n (%) Present Ductal Absent invasion, n (%) Present Serosal Absent invasion, n (%) Present Intrahepatic Absent metastasis, n (%) Present AFP, n (%)
a
PET/CT PET/CT negative, n = 80 positive, n = 67
72 (90.0) 8 (10.0) 17 (21.2) 63 (78.8) 66 (82.5) 14 (17.5) 55 (68.8) 25 (31.2) 43 (53.8) 37 (46.2) 48 (60.0) 32 (40.0) 60 (75.0) 20 (25.0) 78 (97.5) 2 (2.5) 78 (97.5) 2 (2.5) 70 (87.5) 10 (12.5) 39 (48.8) 41 (51.2)
43 (64.2) 24 (35.8) 30 (44.8) 37 (55.2) 34 (50.7) 33 (49.3) 32 (47.8) 35 (52.2) 26 (38.8) 41 (61.2) 22 (32.8) 45 (67.2) 58 (86.6) 9 (13.4) 49 (73.1) 18 (26.9) 63 (94.0) 4 (6.0) 35 (52.2) 32 (47.8) 29 (43.3) 38 (56.7)
P
