Is resectable hepatocellular carcinoma a contraindication to liver transplantation? A novel decision model based on "number of patients needed to transplant" as measure of transplant benefit.

BACKGROUND & AIMS: Number-needed-to-treat is used in assessing the effectiveness of a health-care intervention, and reports the number of patients who need to be treated to prevent one additional bad outcome. Although largely used in medical literature, there are no studies measuring the benefit of liver transplantation (LT) over hepatic resection (HR) for hepatocellular carcinoma (HCC) in terms of "Number of patients needed to transplant (NTT)."

EXCLUSION CRITERIA: Child-Turcotte-Pugh (CTP) Classes B-C, very large (>10 cm) and multi-nodular (>2 nodules) tumours, macroscopic vascular invasion and extra-hepatic metastases.

STUDY POPULATION: 1028 HCC cirrhotic patients from one Eastern (n=441) and two Western (n=587) surgical units. Patient survival observed after HR by proportional hazard regression model was compared to that predicted after LT by the Metroticket calculator. The benefit obtainable from LT compared to resection was analysed in relationship with number of nodules (modelled as ordinal variable: single vs. oligonodular), size of largest nodule (modelled as a continuous variable), presence of microscopic vascular invasion (MVI), and time horizon from surgery (5-year vs. 10-year).

RESULTS: 330 patients were beyond the Milan criteria (32%) and 597 (58%) had MVI. The prevalence of MVI was 52% in patients within Milan criteria and 71% in those beyond (p<0.0001). In the 5-year transplant benefit analysis, nodule size and HCC number were positive predictors of transplant benefit, while MVI had a strong negative impact on NTT. Transplantation performed as an effective therapy (NTT <5) only in oligonodular HCC with largest diameter >3cm (beyond conventional LT criteria) when MVI was absent. The 10-year scenario increased drastically the transplant benefit in all subgroups of resectable patients, and LT became an effective therapy (NTT <5) for all patients without MVI whenever tumor extension and for oligonodular HCC with MVI within conventional LT criteria.

CONCLUSIONS: Based on NTT analysis, the adopted time horizon (5-year vs. 10-year scenario) is the main factor influencing the benefit of LT in patients with resectable HCC and Child A cirrhosis.