Portal de Periódicos da CAPES

Multifaceted perspective of the waiting list for liver transplantation: The value of pharmacokinetic models

2009; Lippincott Williams & Wilkins; Volume: 51; Issue: 1 Linguagem: Inglês

10.1002/hep.23332

1527-3350

Miquel Navasa, Jordi Bruix,

Hepatocellular Carcinoma Treatment and Prognosis

Liver transplantation has become a standard procedure in the management of patients with advanced liver disease, and any referral center in liver diseases should have this intervention within its therapeutic armamentarium. It now constitutes an established field, and any news about transplantation attracts media attention. Victim of its own success, the major problem for a wide application of transplantation is no longer the need to ensure surgical, clinical, and therapeutic skills to manage the selection of candidates, the operation itself, and the postoperative period, but is the shortage of donors. Hence, in almost all health care settings, the number of candidates largely exceeds the available livers to be implanted. Even in countries with high donation rates such as Spain, the excess of candidates induces a steady expansion of the time that patients spend on the waiting list and of their likelihood of death while waiting. As in other areas of health care, there are several actors with separate interests in this complex scenario. HCC, hepatocellular carcinoma; MELD, model for end-stage liver disease. Obviously, the central one is represented by the patients and their referring physicians. As an individual facing a life-threatening disease that might be cured by transplantation, any candidate for transplantation is mostly interested in undergoing transplantation as soon as possible. This expectation is in place even if the prospects for successful long-term survival are highly limited and do not match those required by the current criteria, a circumstance that is especially evident in patients with hepatocellular carcinoma (HCC). Because of the shortage of donors, it is usual policy to exclude from transplantation any patients with an expected suboptimal posttransplant survival. The cutoff is commonly established at less than 50% at 5 years. Nevertheless, it could be argued that if the liver donations would exceed the demand, the requested expected survival could be lowered. The second major player in the field is the transplant team who has to carefully evaluate the candidates and establish if transplant is indicated and if there are no contraindications of any class (comorbidities, failure to control alcohol and tobacco consumption, lack of understanding of the disease etiology by the patient and need to follow specific recommendations, insufficient support, and resources to sustain the demands of the posttransplantation management). Ultimately, the transplant program should have in place a clear and transparent protocol that defines all types of criteria and steps to be fulfilled in a consistent way so that the clinical management of candidates and enlisted patients does not vary arbitrarily through time if there is no robust proof of the need of such a change. One easy example to expose the challenges in the selection process is the decision to have an age limit to enter/exit the waiting list. This limit is in place in several countries but is not homogeneous at all. This demonstrates that there is no major science behind this limit, but just the willingness to use the livers in those subjects in whom the long-term life expectancy and tolerance to the procedure is higher. Should this be a mere chronological assessment or would biological age be better, even though it is not really easy to define? The same heterogeneity in the criteria for transplantation applies to HCC. Most groups adhere to the well-established and successful Milan criteria, but the pressure for expansion is present everywhere. Thus, a noticeable proportion of programs have slightly expanded the limits. In any case, the request of an acceptable outcome in patients who underwent transplantation (again, the suggested 50% at 5 years with as low as possible early mortality) should avoid that this translates into poor service to those patients of more uncertain outcome who could impair the perceived quality of the transplant program (higher mortality, longer hospitalization stay) and put it at risk of failing accreditation. Finally, the third actor comprises the community and the health care administrators. These need to have assurance that the organs provided by the community as a result of donation of any type (brain death, live donation, non beating heart donors…) are used in an equitable way and that the resources allocated to transplantation are efficiently spent. This means that the organs are distributed in a fair system and that they result in the highest possible return of years of life to the community itself. All these forces converge in the liver transplant process, which is governed by Michaelis-Menten kinetics (Fig. 1): the product formation rate (number of patients who undergo transplantation) depends on the enzyme concentration (number of livers from donors) as well as on the substrate concentration (number of patients on the waiting list). Because the number of patients in the waiting list is greater to the numbers of livers from donors, the system becomes saturated. Actually, patients on the waiting list can be considered as the major compartment of a pharmacokinetic model (Fig. 2). The entry rate is defined by the patients enlisted for liver transplantation and the total clearance capacity is the sum of the rate of donation (effective clearance) and the rate of patients excluded while waiting ("toxic" clearance). The first one is limited (donor shortage is the origin of the problem) but the second has no limit and its use is enhanced because the effective clearance is less active. As any pharmacokinetic model predicts, when the entry rate exceeds the clearance capacity, the concentration in the compartment will increase and result in an unacceptably high concentration of the entry entity—patients with and without HCC in this case. Because of the limited number of donors, the effective clearance option will be less efficient and more patients will transit through the "toxic" clearance that would indeed be a toxic pathway resulting in death (Fig. 2). It could be argued that the rate of exclusion may vary because of the severity of the patients enlisted. However, assuming that the criteria for enlistment are fair and properly applied because of the risk of early death or dismal evolution (cancer progression may not translate into immediate death, chronic encephalopathy may also not result in death), it is clear that after some time, the "pharmacokinetic" evolution of the model will result in the "toxic clearance" because of contraindications or death being more active/effective (it is unlimited!) than the "clearance because of transplantation". Major emphasis and publicity has been posed to develop policies to establish priority in the list, which seek to have the sickest undergo transplantation. This supposes to transplant those with major need as defined by risk of immediate death. However, because all patients enlisted are supposed to have a grim life expectancy if they do not receive transplantation, it is no surprise that at a given time point, there will be no way to prevent patients from dying while waiting. At first, the establishment of priority policies (whatever they consist of) will advance the sickest and ameliorate the risk of exclusion. However, patients with severe disease will keep being enlisted and those already enlisted will deteriorate (please recall that they were enlisted because of the risk of disease progression and death in the short term) and they will saturate the effectiveness of the priority selection. There is a third clearance option formed by patients who improve after being enlisted, but if the selection process and enlistment criteria are reliable to select those at need, such a circumstance should be negligible. If not progressing and dying while waiting, the enlistment might have been unnecessary or merely done as a strategy to have a patient already in the system when the need emerges. Priority policies to transplant the sickest were in part developed to eliminate this strategic enlistment and also prompted the decision to impede enlistment of patients with low MELD scores in whom transplantation would even impair their life expectancy. Because MELD is only valid for patients with end-stage liver disease, the other categories of patients (including those with HCC) are given arbitrary MELD points to balance the risk of exclusion while waiting. This transforms the waiting list into a multicompartment pharmacokinetic model in which the goal is to give each category an equitable effective clearance rate (Fig. 2). In that sense, there is significant transplant program heterogeneity in terms of population enlisted and hence, the optimal clearance rate in separate categories in different programs is unlikely to be the same if the entry rate per category is not similar. Liver transplantation and Michaelis-Menten kinetics. The relationship between dosage (number of patients entering the list) and the clearance rate (sum of effective clearance by transplantation and clearance by exclusion and/or death of patients) defines the time to system saturation. Exposure of the waiting list dynamics as a two-compartment pharmacokinetic model. Patients enter the waiting list at a certain rate (k-entry) and are divided into two categories: MELD-served diseases and non–MELD-served diseases (these include HCC among others). The aim of priority policy is that the effective (k-in) and "toxic" (k-out) clearance rate in the two compartments are identical and allow an equal access to liver transplantation by the two populations with a final k-out rate. If the ratio of MELD/non-MELD diseases is not 1, the k1 and k2 rates cannot be the same if the access to transplant has to be equal among categories. Proportions of MELD and non-MELD patients vary among transplant programs and hence, k-in and k-out proposals (meaning priority policy/points) data from different programs cannot be homogeneous among the different populations they have to serve. According to the above considerations, is there any benefit in applying any strategy to prevent enlisted patients from deteriorating and be excluded while waiting? This is the basis of the study by Vitale et al.1 who ran a probability model to estimate the benefits of sorafenib therapy in patients enlisted because of HCC. Prospective trials have shown that this agent delays tumor progression,2 and using the time to progression in the prospective trials, the authors predict to what extent the risk of exclusion is diminished. There is no surprise in the results as the impact of sorafenib in tumor progression is established. However, a major concern prior to accepting this therapy as a valid approach is that we will need to wait for safety data in this specific population. This uncertainty does not affect the waiting period, but rather the immediate postoperative period when active angiogenesis is needed to ensure tissue healing, liver regeneration, and specifically, portal vein and hepatic artery patency. Vitale et al.1 acknowledge this safety need and until these data are gathered within research studies, it is prudent to avoid transplant surgery under the action of antiangiogenics. If it is proven safe, the economic cost of such therapy appears to stay within the current criteria used for policy makers, but this is a political decision, as is any aspect related to health care economy and resource allocation. Clearly, the benefit is relevant for some of the partners but not for all of them. If a patient faces a 6-month wait and his expected time to exclusion or death is 4 months, there is no doubt that the benefit for him exists. However, the impact of this delay in tumor progression and higher likelihood of reaching transplantation is not as relevant for the community. This is a very controversial topic because, to some extent, the aims of the individual patients may not align with the aims of the community that expects to invest the needed money in actions that provide an adequate return to the community itself. Therefore, if the number of donors is limited and the system is saturated according to the pharmacokinetic concepts, it becomes apparent that despite whatever is done while waiting, the number of "saved" individuals cannot possibly be increased. Thus, as a community, there may be no value in keeping an individual patient (or a strata of patients) on the waiting list, because the proportion of "lost while waiting" will not be changed and this makes the investment in keeping patients on the list completely not cost-effective. Furthermore, it could be argued that the life expectancy of those wait-listed is not homogeneously grim if these patients do not undergo transplantation. Patients with end-stage liver disease certainly have a poor short-term outcome, but other subgroups may have a less dismal life expectancy and could still expect to enjoy some years of survival. This is clearly the case of patients with early HCC in compensated cirrhosis. Patients with HCC may benefit from resection or ablation and their life expectancy at 5 years may range between 40% and 70%. Accordingly, the years of life offered by a donated liver (these may range between 8 and 9 years) will provide a higher gain in patients with a lower baseline expected survival: net gain in patients with Child-Pugh class C cirrhosis would exceed 7–8 years, while the gain of the same liver in a patient with HCC would result in a gain of just 3–4 years at most. Patients with HCC account now for almost 30% of the wait-listed patients in some areas, and thus, this consideration is not irrelevant from a community perspective. As stated above, this is not an easy discussion to conduct and it emphasizes that the relevant actors and partners in liver transplantation, and health care in general, are not only the affected patients and physicians, but also the representatives of the community that the health care system should serve. Liver transplantation is probably a very relevant area where a multifaceted approach is certainly needed to avoid a problem of titanic consideration.3 Decisions taken in terms of selection of candidates, priority in the waiting list, and use of the few organs that are available should be evaluated rigorously and continuously by all parties. This would make sure that a limited resource offered by the community returns to the community in an effective and ethical way, thus balancing the individual goals at any level with those of the community.