|
Organ transplant
Patient Information
|
Transplant Update
May 2007
Issue No. 4
View PDF
Dinesh Ranjan, MD
Chief, Transplant Section
In this issue of Transplant Update, Dr. Hoonbae Jeon discusses liver transplantation for hepatocellular carcinoma under the MELD score-based allocation system.
Implemented to prioritize sicker patients for receiving liver transplantation, the MELD allocation system uses serum bilirubin, creatinine and INR in a logarithmic scale to calculate a score that allows the clinician to predict short- and long-term mortality of a patient more objectively based on laboratory tests. An exception is made for hepatocellular carcinoma (HCC) patients who may have a low laboratory MELD score but have high likelihood of short-term mortality due to tumor growth. Since the cirrhotic liver is much more likely to develop HCC, it is important to initiate this screening soon after the diagnosis of cirrhosis.
It is not uncommon for us to encounter cirrhotic patients who have a reasonably preserved biochemical function (low lab MELD score) but have a focus of HCC in their liver on radiological screening. If biochemical dysfunction or decompensation of cirrhosis develops over time, the tumor growth will undoubtedly prevent the consideration of these patients for transplantation. Early detection and screening for HCC is key to survival for these patients.
At UK HealthCare, we encourage early referrals for patients with cirrhosis so we can initiate the screening process, done initially at the time of referral and then subsequently usually on an annual basis. One alternative is to perform the screening locally and keep us informed of the findings. If we identify the HCC focus in the liver, we initiate rapid transplant evaluation and occasionally use loco-regional ablative therapy to retard the growth of the tumor as a bridge to transplantation. Your role as the referring physician is critically important to ensuring we discover HCC in cirrhotic livers early and then address the problem immediately and expeditiously.
Hoonbae Jeon, MD
Traditionally, organ allocation has relied on the recipient’s length of waiting time. Due to unique characteristics of end-stage liver disease patients who do not have any artificial means of maintaining life (such as dialysis for kidney failure or ventricular assist device for heart failure), incidence of waiting list mortality increased with disparity between numbers of transplant candidates and available donor organs. Waiting list mortality due to worsening shortage of organs from deceased donors has focused attention on an improved organ allocation system. This system is based on stratification from medical urgency of transplant candidates rather than a first-come first-served basis. The model for end-stage liver disease (MELD) score, originally developed to predict outcomes of transjugular transhepatic portosystemic shunt procedure (TIPSS) using serum bilirubin, INR and creatinine in a logarithmic calculation, has shown to be useful in ranking transplant candidates according to their probability of death during a defined period. It also has been useful in predicting death independent of etiology and the complications of portal hypertension.
Based on observed superiority of the MELD to the Child-Pugh-Turcotte score as a scale of disease severity and resulted risk of mortality, the new MELD allocation policy was implemented by UNOS on Feb. 27, 2002. In this model, patients are assigned a score that reflects their three-month predicted mortality. In theory, the sickest patient with highest risk of death on the waiting list would get the highest score.
In the previous system with a few tiers of UNOS status, priority status 2B had been given to candidates with early HCC. Despite this priority, waiting time of these candidates was likely to be extended due to poor discriminative capability of the system, which favored the length of waiting time compared to the medical urgency. Because of prolonged waiting time for deceased donor livers for transplantation, these patients were at risk for tumor growth exceeding the acceptable criteria for transplantation. However, in the MELD score-based system, the priority of HCC candidates is maintained by providing extra points in addition to calculated points from the MELD algorithm. Even with increased risk of death from the tumor for relatively preserved liver function in many HCC patients, their MELD score was disproportionately low.
An adjustment was needed to allow for the anticipated death risk from HCC. When the MELD score-based system was implemented, candidates with T1 HCC (single lesion <2cm) were given a MELD score of 24, corresponding to 15 percent three-month mortality. Those with T2 HCC (single lesion 2-5cm, or two to three lesions all <3cm) were assigned a MELD score of 29. Patients with HCC beyond the certain stage would not be given any extra priority than their calculated MELD score based on labs. A study over the first year after implementation of MELD allocation showed the waiting time to deceased donor liver transplantation became significantly shorter and dropout rate from disease progression was reduced. Five-month survival on waiting list during five months was also increased.
The size limit of the tumor for transplant candidacy has been questioned. Yao et al. have demonstrated almost equal prognosis with using their expanded criteria (UCSF criteria) – a single lesion <6.5cm, or three or fewer lesions, a total tumor diameter smaller than 8cm [4]. While using these expanded criteria, dropout rate from the tumor growth can be reduced.
A line between too much or too little priority for HCC patients?
According to national data, a year after implementation of MELD score-based allocation, the proportion of HCC candidates who had transplantation within 30 days was significantly increased (27 percent of stage I, 45 percent of stage II) and 87 percent of candidates received a transplantation within three months. The national rate of transplantation for HCC post-MELD is 21.7 percent for the first year after implementation of MELD score-based allocation, compared to 8.8 percent pre-MELD.
This observation raised concern that HCC candidates were given excessive priority, potentially giving an unfair advantage over other patients without HCC but with comparable MELD scores reflecting the severity of their liver disease. The question whether the prioritization of HCC patients adversely affects sicker cirrhotic patients without HCC has not been completely answered. The issue was addressed in the UNOS/OPTN consensus meeting in January 2003, when the priority MELD score for HCC was decreased from 15 percent to 8 percent (equivalent to MELD score of 20) for patients with T1 tumors and from 30 percent to 15 percent (equivalent to a MELD score of 24) for patients with T2 tumors. This modification took effect on Feb. 27, 2003.
According to the analysis of waiting list outcomes over the next year following implementation of modified MELD prioritization for HCC, which gives reduced extra MELD points to candidates with HCC, there was an increase in probability of waiting more than 90 days and decrease in probability to receive transplantation, but there was no increase in dropout rate due to progression of the disease.
HCC-adjusted MELD scheme also raises concern from the relatively high rate of misdiagnosis of HCC in the liver explant. All transplant programs have to submit the pathologic report of explant liver after transplantation of each patient who was granted extra priority from HCC. However, there is still a significant portion of cases with false positive diagnosis of HCC on pre-transplant imagings [4].
Role of locoregional treatment as a bridge to liver transplantation
Resection has been chosen as the first-line treatment for small HCC’s in patients with no cirrhosis or Child A cirrhosis. Survival after resection has been comparable to transplantation and it is a less expensive treatment without using a donor organ. However, in the case of tumor recurrence or progressive decompensation of hepatic function, salvage transplantation has to be considered. A considerable proportion of patients may survive without recurrence for five years, and among those with recurrence the majority may be eligible for salvage transplantation. On the other hand, it is easy to imagine that salvage transplantation after previous liver resection would be more difficult, associated with more complications and possibly more recurrence.
Chemoembolization has been proposed to play a role in tumor control before transplantation. Dropout rate has been reported as 15-25 percent for six months for patients within Milan/UNOS criteria. Even with chemoembolization, dropout rate still stays around 15 percent. Regardless of bridge treatment before transplantation, dropout rate significantly increases after six months in patients with a tumor larger than 3cm or multiple lesions.
Radiofrequency ablation has been also widely used as a bridge treatment. However, it is very rare to achieve complete tumor necrosis with this modality especially for larger tumors (Figure 1). The value of these bridge treatments has been questioned in recent studies. Survival benefit from either pre-transplant chemoembolization or radiofrequency ablation is hard to determine. However, these bridge treatments have a role in down-staging of tumor mass to within transplantable range. Cases with tumors beyond Milan/UNOS criteria can be submitted to UNOS regional review board to get HCC points approved after size reduction by locoregional treatment (Figure 2). The serial measurement of the tumor before and after the treatment has to be documented and forwarded for the review.
Conclusion
MELD allocation model provides a useful, objective and universal tool for clinicians to estimate risks for clinical decision models to analyze the risk vs. benefit. However, a few questions regarding how much priority should be given over non-HCC patients, true limit of anatomical stage of tumor for transplantation and role of bridge treatment in reduction of dropout and survival benefit after transplantation still remain to be answered. It is also conceivable to develop additional allocation model combined with MELD to place marginal donor livers in HCC patients with relatively better underlying liver cirrhosis but still with high risk of dropout and death.
|
Academics | Service | Research | UK HealthCare | UK A-Z 
