Liver Transplantation for Hepatic Epithelioid Hemangioendothelioma Is Facilitated by Exception Points With Acceptable Long-term Outcomes

Abstract

Background. 

Hepatic epithelioid hemangioendothelioma (HEHE) is a rare vascular tumor with a high mortality rate. HEHE is now a formally recognized indication for exception point priority in the United States under the new National Liver Review Board. The role of liver transplantation (LT) and exception point waitlist priority in the United States for patients with HEHE remains understudied.

Methods. 

This was a retrospective cohort study using the United Network for Organ Sharing transplant database. From February 27, 2002 to January 31, 2018, 131 adults waitlisted for LT with HEHE were identified by free-text entry.

Results. 

Exception point applications were submitted for 91.6% (120/131) of patients. All patients with fully reviewed applications received exception points at least once during waitlisting, and 85% (103/120) upon first submission. Among the 88 patients transplanted, median model for end-stage liver disease score at LT was 7 ((interquartile range [IQR]: 6–11) and waiting time 78.5 days (IQR: 29.5–237.5). Unadjusted post-LT survival of HEHE recipients at 1-, 3-, and 5-years from LT was 88.6%, 78.9%, and 77.2%. Unadjusted post-LT patient and graft survival of HEHE patients was not different from patients with hepatocellular carcinoma within Milan receiving exception point priority (P = 0.08). An increased rate of graft failure due to hepatic artery thrombosis ≤14 days from initial LT was observed in HEHE versus non-HEHE patients (4.6% versus 0.5%).

Conclusions. 

The majority of HEHE recipients receive exception points at a universal approval rate allowing prompt access to deceased donor LT.

INTRODUCTION

Epithelioid hemangioendothelioma is a rare vascular tumor that can originate in the liver, spleen, bone, or lungs, among others. Though considered indolent, primary hepatic epithelioid hemangioendothelioma (HEHE) has a high mortality rate if left untreated, with 1 study reporting only 5% alive at 5 years.¹ There is no consensus approach to the treatment of HEHE because of its low incidence (<1 case/1 000 000) and its heterogeneous clinical course, ranging from rapid progression to disease stability over many years to even spontaneous regression in rare instances.¹ Surgical resection is often not curative nor feasible, as over 80% of patients have bilobar liver involvement at presentation and over one-third develop extrahepatic disease.^2,3 The effectiveness of chemotherapy, transcatheter arterial embolization, and radiotherapy is also poorly understood because of a lack of prospective data and retrospective studies preferentially using such modalities in patients with worse prognosis at the outset.^3,4 Liver transplantation (LT) remains the most commonly employed treatment for HEHE and offers the best outcomes for multifocal disease.¹ In multiple studies, overall survival at 1 and 5 years post-LT reaches 90%–100% and 70%–80%, respectively, with similarly acceptable outcomes in patients with extrahepatic disease.^1,4,5

According to the recently implemented National Liver Review Board (NLRB) by the Organ Procurement and Transplantation Network (OPTN), unresectable HEHE is now a formal indication for model for end-stage liver disease (MELD) exception point priority for LT. According to the NLRB guidance document, patients with HEHE can be considered for exception points after definitive exclusion of hepatic angiosarcoma by biopsy and “the presence of extra-hepatic disease […] should not be an absolute contraindication” to LT.⁶ However, the NLRB guidance statement provides little additional guidance on other important disease predictors known to be associated with post-LT outcomes.^5,7 The objectives of this study were to (1) examine waitlist and post-LT outcomes in a contemporary population-based cohort of patients with HEHE, (2) evaluate the use of exception points for HEHE in the current MELD era, and (3) compare the outcomes of HEHE candidates to those receiving exception priority for other indications.

MATERIALS AND METHODS

This was a retrospective cohort study using national transplant data from the United Network for Organ Sharing (UNOS). Subjects were eligible if they were waitlisted for an initial LT between February 27, 2002 to January 31, 2018, and at least 18 years of age at the time of listing. This start date represents the beginning of the MELD allocation system. Subjects waitlisted for and/or transplanted with multiple organs (eg, liver-kidney) were excluded. HEHE patients were identified based on free-text entry by the transplant center as the diagnosis of record or the reason for exception point request. UNOS diagnostic coding (code 4405) was not used to identify the HEHE cohort, as this also includes patients with hemangiosarcoma and angiosarcoma, which have significantly worse post-LT outcomes.⁸ The concordance of HEHE diagnosis by free-text and by UNOS diagnostic coding was described.

Demographic and clinical characteristics at the time of waitlisting of HEHE candidates were compared with listed patients without HEHE. Exception point practices for HEHE candidates were evaluated. Waitlist outcomes, patient characteristics at LT, and post-LT outcomes of HEHE subjects were compared with those given exception point priority for other indications. Donor organ quality was evaluated using the Donor Risk Index.⁹ Given the high rate of death after retransplantation (reLT) in the HEHE cohort and among LT recipients at large, patient and graft survival were evaluated as a combined outcome.¹⁰

In secondary analyses, unadjusted post-LT patient/graft survival in HEHE recipients was compared with that of patients receiving waitlist exception priority for hepatocellular carcinoma (HCC) within Milan, cholangiocarcinoma or neuroendocrine tumor (NET). Due to their small number, the post-LT outcomes of patients transplanted with hemangiosarcoma or angiosarcoma were described. Posttransplant outcomes were also compared between HEHE patients identified by free-text entry versus UNOS diagnostic coding.

Due to the small size of the cohort, Fisher exact tests and Wilcoxon rank-sum tests were used to describe associations of categorical and continuous variables, respectively. Log rank tests and Kaplan-Meier plots were used to evaluate unadjusted patient/graft survival after initial LT. HEHE was evaluated as a predictor of posttransplant mortality and graft failure using Cox proportional hazards analysis among all patients transplanted with exception points, and unadjusted and adjusted hazard ratios (HRs) were obtained. The following potential confounders were adjusted for in the multivariable model: age, gender, race/ethnicity, receipt of exception points, and laboratory MELD score at LT and days waitlisted.

This study was approved by the Institutional Review Board at the University of Pennsylvania.

RESULTS

Waitlist Characteristics of HEHE Candidates

Between February 27, 2002 to January 31, 2018, 131 individual waitlisted adults with HEHE at 63 centers in the United States were identified by free-text entry, representing 0.1% of all adults waitlisted for initial LT alone during the study period (N = 137, 160). Free-text entry confirmed UNOS diagnosis code 4405 in 94 patients (71.8%). However, 37 candidates waitlisted with HEHE would have been missed had this diagnosis code been used alone.

Patients with HEHE were more likely to be females (70.2% versus 35.4%, P < 0.001) and younger (median age 43 versus 55 y, P < 0.001) than non-HEHE candidates (Table 1). Four HEHE candidates had concurrent intrinsic liver disease: 2 with chronic Hepatitis C virus, 1 with primary sclerosing cholangitis, and 1 with primary biliary cholangitis. The median laboratory MELD score at first listing for HEHE candidates was 7 (interquartile range [IQR] 6–10), and the median serum albumin was in the normal range (3.9 g/dL [IQR 3.4–4.2]).

TABLE 1.: Basic demographic and clinical characteristics of adult HEHE (N = 131) vs non-HEHE (N = 137 153) candidates waitlisted for initial LT alone

Of the 131 patients with EHE, 10 (7.6%) never had an exception point application submitted on their behalf during waitlisting (5 candidates from a single region). Of the 121 patients with at least 1 exception point application, 120 (99.2%) received exception points at least once during waitlisting with 103 (85.1%) having their request approved upon first submission. The sole applicant who did not receive exception points withdrew their only request. At the time of first application submission, the median score requested was 22 MELD points (IQR: 22–24).

Waitlist Outcomes of HEHE Candidates

Of 131 candidates with HEHE waitlisted for initial LT, 88 (67.2%) HEHE candidates underwent LT, 14 (10.7%) were removed because of death or being too sick, 9 (6.9%) were no longer deemed to need LT or refused LT, and 7 (5.3%) were still waitlisted as of January 31, 2018. The remaining 13 HEHE candidates were removed from the waitlist for other unspecified reasons or had missing removal codes.

Among the 10 patients who never had an exception application submitted, 3 were transplanted, 1 died on the waitlist, and 3 were removed from the waitlist because of no longer needing LT. Of the 120 who received exception points at least once during waitlisting, 109 (90.8%) had active exception priority at removal or end of follow-up. Among these, 83 (76.1%) were removed from the waitlist as a result of LT, 11 (10.1%) because of death or becoming too sick, and 1 because of no longer needing LT (0.9%).

Initial LT in HEHE Patients

Overall, 88 patients underwent initial LT for HEHE during the study period: 83 (94.3%) were deceased donor liver transplants (DDLT) and 5 (5.7%) living donor liver transplants (LDLT). Median waiting time before LT was 78.5 days (IQR: 29.5–237.5 d). In 94.3% (83/88) of those transplanted, organ allocation occurred on the basis of active exception points (median match MELD 25, IQR: 22–29). The remaining 5 patients underwent DDLT on the basis of their laboratory MELD score. Clinical characteristics of initial LT recipients with active exception points by HEHE status are shown in Table 2. There was no change in median laboratory MELD score or albumin between waitlisting and LT in recipients with HEHE.

TABLE 2.: Clinical characteristics of adult HEHE vs non-HEHE candidates with active exception points at the time of initial LT

Posttransplant Outcomes in HEHE Patients

After initial LT, 17.1% (15/88) of HEHE recipients died and 8% (7/88) were retransplanted. Of the 15 who died after initial LT, 9 were attributed to recurrent disease and/or malignancy. Unadjusted survival of HEHE recipients at 1-, 3-, and 5-years after initial LT was 88.6%, 78.9%, and 77.2%, respectively. Among patients transplanted with exception points, outcomes after primary LT were not different between HEHE and non-HEHE patients (P = 0.1; Table 2).

Of the 88 initial LT for HEHE, 4 (4.5%) were coded as having graft failure in ≤14 days due to hepatic artery thrombosis (HAT). By comparison, the rate in non-HEHE recipients between 2002 and 2018 was 0.5% (372/78 505). None of the patients with initial graft failure due to HAT in ≤14 days were recipients of split DDLT grafts or LDLT recipients. Of the 7 HEHE recipients who underwent reLT, 5 occurred ≤14 days from initial LT (N = 3 due to HAT, N = 2 due to primary nonfunction or early allograft dysfunction). The 2 remaining reLTs were performed 1 and 13 years after initial LT for late HAT and recurrent HEHE, respectively. Two HEHE recipients were retransplanted a third time, 3 and 36 days after their second LT, both of which as a result of HAT. Among the 7 reLT recipients with HEHE, 5 subsequently died, including both HEHE recipients of 3 LTs. All reLTs occurred at the initial transplanting center, and each occurred at different centers across the United States. With the exception of the 1 patient retransplanted for recurrent HEHE, none of the reLTs occurred in recipients of split DDLT grafts or LDLT recipients.

Among patients transplanted with exception point priority, HEHE as an indication for initial LT was associated with a significantly lower hazard of death/graft failure versus other exception point indications (HR 0.54, 95% confidence interval [CI]: 0.32-0.89; P = 0.017). However, age was identified as a major confounder of this relationship (adjusted HR 0.69, 95% CI: 0.42-1.15; P = 0.2). In secondary analyses, unadjusted survival after initial LT for HEHE recipients was not different compared with recipients with exception priority for HCC within Milan criteria (P = 0.08; Figure 1A) but was superior to those with exception priority for cholangiocarcinoma or NET (P = 0.001; Figure 1B). Posttransplant patient/graft survival was also significantly greater for HEHE recipients identified by free-text compared with those with the UNOS diagnostic code 4405 without free-text entry (P = 0.005; Figure S1, SDC, https://links.lww.com/TP/B818). Of the 8 patients transplanted for hemangiosarcoma or angiosarcoma as identified by free-text, 7 died after a median survival time of 3.5 months, and 1 was lost to follow-up.

FIGURE 1.: Unadjusted survival probability with initial LT of HEHE recipients (N = 88) compared recipients with exception points for HCC within Milan criteria (N = 18 659; A), recipients with exception points for NET (N = 129) or cholangiocarcinoma (CCA; N = 404; B). HCC, hepatocellular carcinoma; HEHE, hepatic epithelioid hemangioendothelioma; LT, liver transplantation; NET, neuroendocrine tumor.

Temporal and Geographic Factors

There were no clear temporal or geographic trends in the waitlisting of candidates with HEHE. Region 5 had the largest number of HEHE candidates waitlisted (26/131, 19.8%) and transplanted (14/88, 15.9%). The 2 largest centers waitlisted 10 HEHE candidates each during the study period. The 88 LTs for HEHE occurred at 50 centers with the largest center performing 6 LTs between 2002 and 2018. A majority of centers (45/63) waitlisted only 1–2 patients with HEHE, and 40/50 centers transplanted 1–2 patients during this 15-year period.

DISCUSSION

Since 2002, transplant centers may apply for exception points for conditions in which survival is not accurately estimated by the MELD score. Prior research has shown that exception points have been granted inconsistently, leading to the implementation of the NLRB by the OPTN earlier this year, and HEHE is now one of the recognized indications for exception point priority on the LT waitlist.^11-14 This study demonstrates that a large majority of LTs for HEHE occur as a result of exception points, while LDLT is seldom pursued for this condition. The approval rate was universal among patients with fully reviewed applications, and 85% of patients received exception points upon first submission. Given the heterogeneous nature of this rare disease, these findings highlight the lack of available evidence needed to comprehensively assess the need for expedited LT in patients with HEHE. The new NLRB guidance statement unfortunately remains vague, and it is uncertain whether future candidates with HEHE will be considered with greater scrutiny.

When faced with an extremely rare condition, physician decision-making relies on prior published experience with multicenter patient registries and/or databases being the most reliable sources of information. This research adds to the limited existing literature on the waitlist and transplant outcomes of candidates with HEHE as well as the role of exception points in gaining access to LT in the United States. A majority of centers have waitlisted only 1–2 patients with HEHE in the last 15 years, thus few physicians have experience with the role of LT for this condition. The transplant community would greatly benefit from designating a select group of centers with the most experience with HEHE as centers of excellence, facilitating not only future research in this area but also clinical guidance.

Because of the small number of patients without exception points, the outcomes of exception versus nonexception point HEHE patients could not be compared. However, given that exception points provide an accelerated pathway to transplant, this study did explore the post-LT outcomes of HEHE patients in the context of those receiving exception points for other hepatobiliary tumors. These exploratory analyses demonstrated that unadjusted post-LT survival in HEHE patients was comparable with that of patients with HCC within Milan criteria and superior to patients receiving exception points for cholangiocarcinoma or NET. Thus, despite the universal exception point approval rate observed, the allocation of additional waitlist priority for this condition appears favorable in most patients. It is unknown why exception point priority was not sought in 10 of the 131 waitlisted HEHE patients, particularly as significant hepatic synthetic dysfunction is rarely present. Interestingly, none of these patients underwent LDLT, 5 of 10 were from a single region, and 3 of 10 no longer needed LT altogether. Potential explanations include (1) inadequate center experience with the exception point system for nonstandard conditions, (2) preexisting geographic variability in exception point approval rates for other nonstandard conditions, (3) uncertainty regarding transplant benefit and/or timing given clinical circumstances, and (4) the desire to pursue additional treatment modalities before potential LT.

Because of the variable behavior of HEHE, identifying optimal timing for LT is frequently challenging. The European experience of LT for HEHE suggests that waiting time <120 days predicts worse post-LT outcomes, yet in the United States the median pre-LT waiting time was only 79 days.⁵ Patients with HEHE may warrant a predefined period of observation before LT, similar to the new exception point policy for HCC.¹⁵ In addition, 28% of HEHE patients in the study by Lai et al received neoadjuvant treatment, and 40% underwent supplementary surgery at time of LT (eg, lung resection). Neither of these issues are addressed in the OPTN guidance document nor are other known predictors of worse outcomes post-LT.^5,7 Lastly, transplant centers should encourage the use of LDLT for HEHE so as to avoid potentially disadvantaging waitlisted candidates in whom the survival benefit of LT is clearly established.

In this study, it was observed that HAT may be more common in LT recipients with HEHE, which is a novel finding. Approximately 4.5% of patients transplanted for HEHE developed graft failure due to HAT within 14 days of initial LT compared with 0.5% of non-HEHE patients. Early or late HAT were also the most common indications for reLT among patients with HEHE. Moreover, none of the identified cases of HAT occurred in recipients of split graft DDLT or LDLT. It should be noted that the numbers of graft failures and reLTs for HAT in this study were small, and thus these data cannot be used in isolation. Future research is needed to further establish whether differences in the risk of post-LT HAT truly exist in the HEHE population.

As described, there are inherent limitations to this small study. The UNOS database does not contain details of disease extent or treatment history. Utilizing the free-text entry to identify HEHE patients in lieu of the UNOS diagnostic coding likely increased the positive predictive value, though some HEHE patients may have been missed as a result. The unadjusted survival of LT recipients identified by UNOS coding without free-text entry was markedly inferior to the HEHE study cohort, likely due to the inclusion of patients with angiosarcoma. Biopsy is now mandatory to be eligible for exception points, as nuclear expression of a CAMTA1 mutation reliably differentiates HEHE from other vascular tumors, though this was not the case before 2019.¹⁶

In conclusion, LT for HEHE offers acceptable outcomes in the MELD allocation era. However, given the known disease heterogeneity, greater detail is needed for future NLRB exception point guidance. LDLT should also be encouraged to avoid the possibility of disadvantaging non-HEHE candidates on the waitlist. The possible increased frequency of HAT in the HEHE recipient population is a new finding that requires further investigation. Given the rare incidence of this disease overall, a US-based registry and/or designated centers of excellence for HEHE would be a great benefit for future research and clinical decision-making.