Liver Transplantation for Malignancies

CMS NCA document | source_status=Closed | review_type=New | public_comment_open=False | document_id=CAG-00091N

Our review of the scientific evidence indicates that liver transplantation in patients with HCC is reasonable and necessary in selected patients. Specifically, we evaluated patients for who transplant is treatement for HCC and patients transplanted for other reasons who are found to have HCC. There are a number of retrospective comparison studies that demonstrate that both actuarial survival and recurrence-free survival in groups of patients with HCC treated with transplantation achieve results comparable to non-malignant transplant cases. Generally, we would prefer to have prospective comparative studies on which to base our determination. However , given the life-threatening nature of HCC for patients who are not eligible for liver resection, we believe there may be legitimate ethical concerns surrounding randomization of patients for a prospective trial of this procedure. While the existing retrospective studies may have allowed for some selection bias in the form of the HCC patients undergoing transplant, some studies adjusted for this possibility. In addition, the studies clearly demonstrate that properly selected HCC patients have good long-term outcome following transplant. Thus, we believe that there is sufficient evidence in this case to support a NCD for adult liver transplantation for HCC in specific circumstances. Through the literature we reviewed, several patient characteristics were found to strongly predict outcome in liver transplant patients. Thus, this decision memorandum announces Medicare's intention to make a national coverage decision for liver transplantation for patients with HCC only under the following circumstances: The patient is not a liver resection candidate; The patient's tumor(s) is less than or equal to 5 cm in diameter; There is no macrovascular involvement; and, There is no identifiable extrahepatic spread of tumor to surrounding lymph nodes, lungs, abdominal organs or bone. Most of the literature we reviewed explicitly stated that liver transplant should be reserved for those patients for whom liver resection is not an option primarily due to poor liver function or location of the tumor. Unlike patients with liver resection, patients receiving liver transplantation are on a lifetime regimen of immunosuppression, which presents ongoing health risks, such as increased risk of infections. Because of these risks associated with liver transplantation we do not believe transplantation is reasonable and necessary when liver resection is likely to produce a comparable result. Virtually all of the literature we reviewed identified tumor size greater than 5 cm as an adverse risk factor for liver transplantation for HCC. While many of the researchers noted an association between tumor size and negative outcomes, Klintmalm, Figueras et al.(1997) and Iwatsuki et al.found tumor size as a significant variable in predicting outcome. In their most recent research, Iwatsuki et al. determined that a patient with tumors greater than 5 cm is 6.7 times more likely to have a low chance of survival at 5 years. The UNOS criteria consider tumor size in determining the status of patients on the wait list. Patients with a single tumor less than 5 cm or 3 or fewer tumors all less than 3 cm may be considered as a Status 2B candidate on the list. Patients with larger tumors may only be considered as a Status 3 candidate and will have a significantly reduced chance of being allocated a donated liver. Given the evidence linking tumor size with poor outcomes, we do not believe that it is reasonable or necessary to perform liver transplants on patients with tumors greater than 5 cm. From the evidence we reviewed in analyzing liver transplantation for HCC, the most consistently discussed significant adverse prognostic factors were macrovascular invasion and extrahepatic spread. Figueras et al., Iwatsuki et al. and Klintmalm all found these factors to be a significant negative influence on the likelihood of survival especially for disease free survival. Iwatsuki et al. found that transplantation of patients with macovascular tumor invasion increased their risk of poor outcome by a factor of 15 times. Hepatocellular carcinoma that invades the vascular system is more likely to recur and effect the grafted liver. Similarly, the UNOS criteria for Status 2B include a condition that patients be evaluated to "rule out any macrovascular involvement." We have concluded that the evidence supports a determination that liver transplantation for HCC with macrovascular invasion is not reasonable and necessary given the magnitude of the risks for recurrence. The researchers likewise found extrahepatic spread of HCC significantly affected the outcome of liver transplantation as a means of curing HCC. Iwatsuki et al., Klintmalm, Mazzaferro et al. and Yamamoto et al. all identified extrahepatic spread as a factor that negatively effected the outcome of liver transplantation. Iwatsuki et al. found recurrence of cancer in these patients to be universal within 2 years of transplantation . Klintmalm identified extrahepatic spread as an independent variable in his regression analysis and Mazzaferro et al. identified this as the only significant factor in his analysis. UNOS status 2B criteria specify that extrahepatic spread is to be ruled out. Given the strong evidence for rapid recurrence of cancer in patients with extrahepatic spread, we concluded that the likelihood of risks that exceed benefits for patients in this category is high. Therefore, we have concluded that liver transplantation where there is evidence of extrahepatic spread is not reasonable and necessary. Several researchers have identified tumor number and staging as adverse factors. Further, the UNOS status 2 B criteria are limited to patients with Stage I or Stage II tumors in accordance with the Tumor-Node-Metastasis (TNM) classification system. This results in Status 2 B classification only to patients with 3 or fewer nodules. We considered use of tumor numbers and staging as independent factors in establishing our conditions for coverage in the decision memorandum. However, we found that many of the researchers did not study or find any negative relationship between tumor number and outcome. Of those researchers that did study this variable, all found a negative association, but failed to identify it as being statistically significant. More researchers studied tumor staging as a variable. Similar to tumor number, the researchers consistently found a negative association of outcome with higher staging. That is, as the stage of the tumor increased, the disease-free survival of the patients decreased. But only one researcher, Klintmalm, in his 1998 work, found this association to be statistically significant. Given the high likelihood of occurrence of other significant adverse factors, such as increased tumor size, macrovasularization and extrahepatic spread, to occur in patients with higher stage tumors, it is possible that these other factors are the cause of the negative association of increased staging with outcomes. Therefore, we have not included these factors in our conditions for coverage in the decision memorandum. In addition, several researchers found a negative association with outcomes for bilobar distribution and microvascularization. However, the only factor identified as statistically significant was bilobar distribution in the Iwatsuki et al. study in 1991. In Iwatsuki's later research, he empirically weighted these risk factors as 3.1 for bilobar distribution and 4.4 for microvascularization. We note that the UNOS status 2 B criteria do not consider either of these factors. Further, identification of microvascularization prior to transplantation requires the use of a biopsy procedure that, in and of itself, may be an adverse risk factor in patients with HCC in that the procedure may seed the tumor. Thus, we concluded that a link between these factors and poor outcomes is unlikely. Consequently, we did not exclude patients with these conditions from our conditions for coverage in the decision memorandum. In summary, we have decided to issue this decision memorandum announcing our intention to revise the NCD relating to liver transplantation for malignancies. We will issue a revision to the Coverage Issues Manual stating that liver transplantation may be covered for patients with HCC when the following conditions are met: The patient is not a liver resection candidate; The patient's tumor(s) is less than or equal to 5 cm in diameter; There is no macrovascular involvement; and There is no identifiable extrahepatic spread of tumor to surrounding lymph nodes, lungs, abdominal organs or bone. Liver transplantation for other malignancies continues to be excluded from coverage at this time. We will, however, seek a technology assessment on these other types of malignancies. HCFA's coverage policies are updated as new scientific information becomes available. We welcome comments on this document now or in the future. In addiiton, HCA will accept a request for reconsideration of this decision memorandum or any NCD if there is felt to be a Federal misinterpretation of existing evidence or if new evidence is provided that may alter the conclusion of this assessment.