One of the major tenets of fair organ allocation is the equitable access and distribution of organs to candidates regardless of their disease. Because of the great scarcity of available organs, the utilitarianism of the final rule as a measure of normative ethics dictates that organs should be allocated to those patients who will maximize the utility of the grafts in terms of survival after transplantation and have the greatest risk of wait-list dropout, whether this dropout is due to death or other reasons (eg, tumor progression).
Patients with hepatocellular carcinoma (HCC) constitute a group of patients who would be poorly served by pure Model for End-Stage Liver Disease (MELD) allocation. In the absence of MELD exception points, patients with HCC would have an increased risk of wait-list dropout, generally because of tumor progression, even though they have a relatively low risk of wait-list mortality due to liver disease. Accordingly, it was determined that the risk of tumor progression, rather than the 3-month risk of death, should be used to adjust the MELD score for these patients. Indeed, the adoption of a risk-based allocation system using MELD exception points has lowered wait-list mortality. With HCC now representing the primary indication for liver transplantation for more than a quarter of all transplants, a careful examination of the prioritization of these patients is mandatory.
MELD exception points for HCC were originally granted to patients whose tumor burden was within the so-called Milan criteria (stage T1 or T2). However, single-center data, registry data, and Markov modeling have all indicated that the initial judgments overestimated the risks of tumor progression.[4, 5] Using such data, policymakers have continued to refine the prioritization system for HCC patients. First, it was concluded that patients with stage T1 tumors (single lesion < 2 cm) would no longer be assigned exception points because virtually none of these candidates experienced wait-list dropout within 1 year. Second, the number of initial exception points granted to patients with T2 tumors was decreased to 22 points (equivalent to a 15% risk of mortality at 3 months).[6, 7] Additional revisions of the HCC allocation policy are ongoing.
Despite these efforts, data continue to suggest that with substantially lower wait-list dropout rates and high transplant rates, HCC patients remain at a marked advantage in comparison with non-HCC patients.[8, 9]
In this issue of Liver Transplantation, Mehta et al. identify a specific cohort of HCC patients with T2 tumors whose risk of wait-list dropout is very low. The authors performed a retrospective, single-center analysis of adult HCC patients listed for liver transplantation with MELD exception points. They used a competing risk analysis to calculate cumulative incidence rates of dropout from the wait list between the years 2005 and 2011. The authors found that predictors of dropout due to tumor progression or death included the presence of a 3 to 5 cm solitary tumor versus a <3-cm solitary tumor, the presence of more than 1 tumor, the lack of a complete response to locoregional therapy (LRT), and an alpha-fetoprotein level > 20 ng/mL versus an alpha-fetoprotein level ≤ 20 ng/mL. Therefore, they have identified a subgroup meeting all 3 criteria (a 2 to 3 cm solitary tumor, a complete response to first LRT, and an alpha-fetoprotein level < 20 ng/mL after first LRT) who have a substantially lower risk of wait-list dropout.
In a setting in which geographic disparities significantly affect liver graft allocation, decreasing the prioritization of all HCC patients across the board nationally, as currently proposed, would likely disproportionately compromise transplantation rates for HCC patients in certain geographic regions. Therefore, the identification of a subgroup that could be safely given less priority would of great value. Indeed, the authors conclude that the adoption of these criteria could have significant implications for organ allocation policy for HCC patients.
These data further complement other reports. For example, Merani et al. showed that despite very different risks of wait-list dropout, all patients with T2 HCCs have equitable access to liver grafts, and Cucchetti et al. demonstrated that dropout rates are significantly affected by the MELD score, the tumor stage, and the response to LRT.
In our opinion, the current study represents a step toward the better allocation of liver grafts for HCC patients. However, it also raises several questions. First, in this study, a substantial proportion of listed T2 HCC patients were excluded. These were primarily patients who underwent down-staging to meet T2 criteria. Thus, it should be emphasized that these criteria should be applied only to those with T2 tumors at presentation. Second, and perhaps most importantly, posttransplant outcomes were not studied. This is particularly important should the organ allocation policy deprioritize the identified subgroup. It should be pointed out that the clinical characteristics of HCC patients identified in this study as having a greater risk of wait-list mortality/dropout are the same characteristics that have been generally associated with poor transplant outcomes. For example, posttransplant outcomes for HCC patients are affected by the tumor size and alpha-fetoprotein levels. It may very well be that these patients who could be restricted from receiving MELD exception points are the ones who enjoy optimal posttransplant outcomes. Indeed, Cucchetti et al. showed that the higher the probability of wait-list dropout is (as predicted by the HCC-MELD algorithm), the worse the survival is after LT. The final rule does not solely speak of pretransplant risk; it is also aimed at maximizing posttransplant outcomes. Thus, future studies will need to be focused on the posttransplant outcomes of the identified low-risk subgroup versus the remaining HCC patients and, ideally, non-HCC patients altogether.
The adoption of such a policy for restricting MELD exception points to those with a greater risk of tumor progression or wait-list mortality could have 1 positive benefit. The largest number of living donor liver transplants in the United States was performed in the year before the initiation of MELD allocation. Although the significant drop-off in the number of living donor liver transplants has generally been associated with the publicity surrounding a donor death in New York City, one could make the argument that because patients with HCC are so highly prioritized with the change to MELD allocation, living donor liver transplantation is less likely to be used for such patients. Living donor liver transplantation may be an optimal way for providing transplants to patients with HCC who could be deprioritized by such a policy.
There are 2 main lessons to be learned here. First, there exists a population of patients with a low dropout risk who are currently given excessive prioritization on the transplant list. Second, the equitable allocation of deceased donor organs requires not only an examination of wait-list outcomes but also the continual examination of posttransplant outcomes as we go forward. Future research will be needed to describe the risk of malignancy recurrence in the HCC patients still receiving priority. The tide may be changed from simply using the Milan criteria to identify candidates for liver transplantation to using a more nuanced tumor biology–driven approach.
In summary, this study by Mehta et al. shows that the current allocation policy may need a revision in terms of certain cohorts that are unjustly advantaged. The authors have identified specific criteria for optimizing these algorithms in concordance with the final rule. Of course, these data will need to be confirmed by larger, multicenter studies, which will also need to be focused on posttransplant outcomes.