Incidental Cholangiocarcinoma is Associated with Poor Outcome in Patients Transplanted for Hepatocellular Carcinoma

Background: Hepatocellular carcinoma (HCC) is a common malignancy in Asia for which orthotopic liver transplantation (OLT) offers curative treatment in selected patients. Incidental HCC-cholangiocarcinoma (HCC-CLC) has been associated with poor OLT outcomes. Aims: To examine the prevalence of incidental HCC-CLC in liver explants from patients undergoing liver transplantation for HCC in our center and to examine its association with post-transplant disease recurrence and mortality. Methods: Medical records of all liver transplant patients in our center were reviewed. The presence of HCC-CLC in liver explants was reviewed by experienced pathologists. Factors associated with recurrence of HCC post OLT and post-transplant mortality were analyzed. Survival in patients with incidental HCC-CLC was compared against those without. Results: A total of 54 transplants were performed during the study period, of which 24 were transplanted for HCC. Mean age was 57 ± 9 years and 87.5% were male. Incidental HCC-CLC was documented in the explants of 4 patients (16.7%). Of these, 2 developed HCC recurrence and 3 died within 24 months. Over a median follow-up of 36 months, 3(12.5%) developed HCC recurrence and 8(33%) died. Factors associated with HCC recurrence and death were number and size of HCC lesions, microvascular invasion and incidental HCC-CLC. Age and AFP level were associated with recurrence but not with mortality. Mean survival of patients with HCC-CLC was poorer compared to those without. Conclusions: Incidental HCC-CLC is not uncommon in OLT for HCC and is associated with tumor recurrence and poor survival. OLT between 10-20% despite strict adherence to the proposed criteria Recognized factors associated with recurrence of HCC in transplanted livers include tumor size beyond Milan and UCSF vascular invasion on explanted liver, poor differentiation of HCC on histopathology, previous hepatectomy and pretransplant alpha-fetoprotein (AFP) level ng/ml Post-transplant recurrence of HCC is associated with reduced Several recent studies have observed that unrecognized mixed HCC-cholangiocarcinoma are associated with poor post-transplant outcomes with increased recurrence rates and reduced survival The aim of our study was to examine the prevalence of incidental HCC-cholangiocarcinoma in liver explants from patients undergoing liver transplantation for HCC in our center and to examine its association with post-transplant disease recurrence and mortality.


Introduction
Hepatocellular carcinoma (HCC) is the third commonest cause of cancer-related death worldwide [1]. Orthotopic liver transplantation (OLT) can be regarded as the best treatment modality where both tumor as well as diseased liver are removed and replaced with a new, disease-free liver. In OLT for HCC, United Network for Organ Sharing (UNOS) system for allocating organs in the United States and many other liver transplant centers in the world select patients based on Milan criteria i.e. single HCC lesion ≤5 cm or 2-3 lesions ≤3 cm each with no vascular invasion or distant metastasis [2]. Another commonly adopted selection criteria is that proposed by the University of California San Francisco (UCSF) that is a slight expansion of Milan criteria to benefit more patients with HCC for OLT (single HCC lesion ≤6.5 cm and 2-3 lesions ≤4.5 cm each and cumulative diameter not more than 8 cm with no local or systemic spread) [3][4][5]. Proper selection of HCC patients with the Milan and UCSF criteria has improved the recurrence-free survival and survival rates at 5-years post liver transplant to approximately 83% and 75% respectively.
Several recent studies have observed that unrecognized mixed HCC-cholangiocarcinoma are associated with poor post-transplant outcomes with increased recurrence rates and reduced survival [16][17][18][19][20][21][22]. The aim of our study was to examine the prevalence of incidental HCC-cholangiocarcinoma in liver explants from patients undergoing liver transplantation for HCC in our center and to examine its association with post-transplant disease recurrence and mortality.

Methods
A retrospective review was performed from the medical records of all patients who underwent liver transplantation at the Singapore General Hospital (SGH) since the inception of the liver transplant program in February 2006 to June 2012. SGH is the largest tertiary hospital and one of two government-funded liver transplant centers in Singapore. Majority of OLT in SGH are deceased donor liver transplantations (DDLT). Our program adopts the UCSF selection criteria to enroll HCC patients for OLT (single tumor <6.5 cm, maximum of 3 total tumors with none >4.5 cm, and a cumulative tumor size <8 cm). Patients who did not fulfill criteria for liver transplant were excluded from the study.
The socio-demographic data of patients transplanted for HCC were evaluated (age, race, gender, etiology of liver disease and AFP level). Pre-transplant imaging that included ultrasound, quadriphasic CT scan and contrast enhanced dynamic MRI scan were reviewed for number of lesions, size of individual and largest lesion, cumulative size of the lesions, vascular invasion, portal vein thrombosis, tumor thrombosis and evidence of distant metastasis. UCSF criteria was used to enroll the patients for transplant. These features were compared with operative findings, gross and histo-pathological evaluation of the explanted liver. Presence of mixed HCC-cholangiocarcinoma in the explanted livers was evaluated by experienced pathologists. Treatment of HCC prior to transplant with liver resection and loco-regional therapeutic procedures such as percutaneous ethanol injection (PEI), trans-arterial chemoembolization (TACE), trans-arterial embolization (TAE), radiofrequency ablation (RFA) were documented.
After OLT, patients were prospectively monitored for recurrence of disease. Post-transplant HCC surveillance was performed via 3monthly AFP and CT scan or ultrasound abdomen. In cases of HCC recurrence, the site of recurrence and time interval of recurrence after OLT were recorded. Type of immunosuppressant medications used and in cases with underlying viral etiology, hepatitis B DNA and hepatitis C RNA quantitative levels were documented before and after transplantation. In patients who died, the cause of death and time interval from OLT was recorded.
IBM SPSS Statistics (version 19) was used for data analyses. Descriptive analyses of demographic variables were performed. Continuous variables were expressed as mean and standard deviations (SD) and categorical variables as proportions. Continuous variables were compared using Student t test and categorical variables were compared by chi-square or Fisher exact test and 95% confidence intervals (CI) was calculated for each association. All p-values were two sided and considered as statistically significant if p <0.05. Survival comparisons were performed using Kaplan Meier analysis and compared using the log-rank association.

Results
A total of 54 patients underwent liver transplantation at our center during the period of study. Of these, 24 (44.4%) were performed for HCC as the primary indication. In patients who underwent OLT for HCC, 19 (79.2%) had DDLT and 5 (20.8%) had living donor liver transplantation (LDLT). Mean age of patients was 57 ± 9 years and 87.5% were male. Underlying liver diseases leading to HCC were hepatitis B cirrhosis in 16 (66.7%), hepatitis C cirrhosis in 4 (16.6%) and cryptogenic cirrhosis in 4 (16.7%). None of the patients had underlying primary sclerosing cholangitis. Mean MELD score was a HCC-assigned score of 15 as the indication of transplantation was HCC rather than severity of disease in majority of these cases (Table  1).

Demographics N=24
Age in years (mean ± SD) 57 ± 9 Male gender 21 (87.5%)  Of the 24 patients who underwent OLT for HCC, 3 (12.5%) patients had recurrence of HCC over a median follow-up period of 36 months. The first had early recurrence at 6 months post-OLT. This patient had multiple lesions (beyond UCSF criteria) and microvascular invasion of tumor into portal vein in the explanted liver which were not apparent on pre-transplant imaging. The patient was treated with loco-regional ablative therapy but developed multiple recurrences and eventually died 4.4 years after OLT. In the second case, recurrence occurred at 18 months after OLT and the patient died within 2 years with lung metastasis. Histological review of the explanted liver revealed multiple liver nodules, evidence of micro-vascular invasion of HCC and an associated 3.3 cm HCC-CLC nodule. The third patient had recurrence of HCC at 2 years. This patient had several liver nodules that were interpreted as indeterminate at pre-transplant imaging. However, on histological examination of the explanted liver, these nodules were confirmed to be HCC. There were a total of five HCC lesions in the explanted liver, which was beyond Milan and UCSF criteria. In this case there was no evidence of microvascular invasion of the HCC but a 2 cm nodule of HCC-CLC was detected. This patient had recurrence of HCC in regional lymph nodes at the porta hepatis and peritoneal cavity and died due to complications of HCC. On univariate analysis, factors associated with HCC recurrence included presence of microvascular invasion on liver explant (67% vs. 4.8%, p=0.03), incidental HCC-CLC (67% vs. 9.5%, p=0.06), age (45 ± 19 vs. 58 ± 6 years, p=0.02), number of HCC lesions on liver explant (5.7 ± 1.1 vs. 2.4 ± 2.7, p=0.06) and AFP level (846 ± 911 vs. 164 ± 376 IU/ml, p=0.02). Disease-free survival at one, three and five years was 96%, 87% and 87% respectively.

Hepatitis
A total of 8 patients died over a median follow-up period of 36 months (range 1 to 68 months). Two (8.3%) patients died from HCC recurrence. The other six demised from sepsis with multi-organ dysfunction syndrome (MODS), primary non-function of liver graft, portal vein thrombosis and hepatic artery thrombosis (Table 3). On univariate analysis, factors associated with mortality in patients transplanted for HCC include >3 HCC lesions, largest HCC lesion >6.5 cm, LDLT, presence of microvascular invasion and incidental HCC-CLC in the liver explant (Tables 3 and 4). Pre-transplant AFP>1000 ug/L was not associated with mortality in our study. Multivariable analysis could not be meaningfully calculated due to small sample size. Overall median survival at one, three and five years was 79%, 71% and 71% respectively. None of the patients was subjected to re-transplantation and there was no living donor mortality. Patients with HCC-CLC demonstrated a trend for reduced survival compared to those without HCC-CLC (mean survival 19.2 Vs. 52.0 months, p=0.06 by log-rank comparison) (Figure 1).

Discussion
Orthotopic liver transplantation is a standard of care in properly selected patients with hepatocellular carcinoma. Milan and UCSF criteria are currently used to enlist the patients with HCC for OLT in majority of transplant centers [23]. Nevertheless, despite strict adherence to these criteria, recurrence of HCC is still reported in 10 to 20% of the patients after OLT [6,7]. These figures highlight that apart from size, number and metastasis of the HCC, other factors may also play role in the recurrence of HCC post-OLT.
In this review of liver transplantation for HCC from a single center, we observed a prevalence of incidental HCC-CLC of 16.7% in the liver explants. Despite being a low volume transplant center, we encountered a relatively high prevalence of HCC-CLC. Sapisochin et al. reported incidental cholangiocarcinoma in the explants of 10 of 302 patients (3.3%) who underwent liver transplant for HCC in a large North American center [18]. A small study from Saudi Arabia reported incidental cholangiocarcinoma in 2 out of 23 patients (8.7%) who had liver transplant for HCC [19]. In comparison, Vallin et al. [20] and Patkowski et al. [21] reported lower HCC-CLC prevalence rates of 1% and 0.8% respectively. However in these latter studies, the low prevalence rates were observed from a denominator of all liver transplants performed for both HCC and non-HCC indications.
Incidental HCC-CLC is associated with recurrence of HCC post OLT [16]. In our series, two out of three cases of HCC recurrence had incidental HCC-CLC in the explanted liver. In both cases, the recurrences were proven on biopsy to be HCC without histological evidence of associated cholangiocarcinoma. In addition, AFP was elevated in both cases of HCC recurrence. This interesting association has been reported in other studies demonstrating increased risk of HCC recurrence and more aggressive disease in patients with coexisting cholangiocarcinoma and HCC [17]. Our results are consistent with existing literature on the association of incidental HCC-CLC and increased risk of HCC recurrence. Sapisochin and colleagues have described a tumor recurrence rate of 57% in liver transplant patients with incidental HCC-CLC [18]. A French multicenter study [20] similarly reported a post-transplant recurrence rate of 50% amongst 10 cases of incidental HCC-CLC whereas the Polish study [21] reported recurrence rates of 73% in liver transplant patients with incidental cholangiocarcinoma. The reason for this observation remains elusive; further studies are required to determine if there is a direct effect of the presence of concurrent cholangiocarcinoma on the recurrence of HCC.
The results from our study also demonstrate a trend towards reduced survival after OLT in HCC patients with a concurrent focus of cholangiocarcinoma. Average survival of patients with HCC-CLC was only 19 months compared to 52 months in OLT patients without HCC-CLC. Although the log-rank comparison did not reach statistical significance due to the small number of subjects, the poorer prognosis observed in HCC-CLC patients is consistent with reports from other centers [18][19][20][21]. In fact, post-transplantation cholangiocarcinoma recurrence has been reported to have a dismal recurrence-free survival of 0% at 3 years [21].
Other factors associated with recurrence of HCC in this study on univariate analysis were size of largest lesion more than 6.5 cm, number of lesions more than three and micro vascular invasion on explanted liver. These findings were consistent with previously reported risk factors for recurrence of HCC after OLT [10][11][12][13][14][15]. In this series, our numbers were too low to calculate meaningful multivariate analysis 95% confidence interval. Many studies have also suggested that aggressiveness of HCC (as evident by poor differentiation of tumor on histopathology) are associated with increased recurrence. [10][11][12][13][14][15]. However, all 3 cases of recurrence in our study had welldifferentiated HCC and this difference may be due to small number of HCC recurrence cases in this study. Contrary to reports that a pretransplant alpha-fetoprotein (AFP) level of 1000 ng/ml or more is a predictor of recurrence, all of our cases of HCC recurrence had AFP level ≤1000 ng/ml [24,25]. In this series, all the recurrences of HCC occurred within two years of transplantation and none of the patients transplanted for HCC had recurrence beyond two years and similar observation was reported in other studies [10][11][12][13][14][15]. Therefore close surveillance for recurrence is recommended in the initial post transplantation period in order to initiate effective therapy earlier.
In our study, the median and disease-free survival at one and five years were 79%, 71%, and 96%, 88% respectively, which are similar to other reviews [10][11][12][13][14][15]. Pre-transplant evaluation of the HCC was by ultrasound, quadriphasic CT scan and thorax and dynamic contrast enhanced MRI scan. However despite advances in imaging techniques, there remains a discrepancy in pre-operative staging of tumor and actual number of tumor lesions in the explanted liver. In all the three patients with recurrence of HCC in this study, pre-operative imaging had underestimated the number, size and vascular invasion of the HCC. Disparity between pre-operative imaging and disease staging in the explanted liver has been reported in up to 20% of cases by Decaens et al. [5]; the authors suggest caution in expanding selection criteria based upon tumor size and number on pre-transplant imaging.
The immunosuppressant regime used in majority of our patients was a combination of prednisolone, tacrolimus and mycophenolate mofetyl (MMF). Sirolimus was used in only one patient because of significant renal insufficiency and low cell counts. Several studies in the literature has suggested that a mTOR inhibitor like sirolimus may have anti-neoplastic activity against HCC recurrence [26,27].
We acknowledge that the interpretation of the findings described in this retrospective single-center study is limited due to the small sample size and require further validation in a larger multi-center prospective study.
In conclusion, we have demonstrated that incidental HCC-CLC is not uncommon in OLT for HCC and is associated with tumor recurrence and poor survival. The role of pre-emptive treatment with combination of chemotherapy and radiotherapy in an attempt to reduce rate of HCC recurrence remains to be explored.

Compliance with Ethical Requirements
All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 [5]. In this study, data was obtained from medical chart review and computer based recorded Information so patients' consent was not required. No identifying information about patients is included in the article.