ADDICTOlogical Intervention in LIVEr Transplantation Recipients

Liver transplantation (LT) is the only curative option for end-stage liver disease and unresectable hepatocellular carcinoma (HCC) without extrahepatic spread. Alcohol Associated Liver Disease (AALD) has become the most common indication for liver transplantation (LT) in many Western countries.In France, AALD accounts for at least 40% of all LT, between decompensated cirrhosis and HCC, which represents more than 500 patients each year.

One, five and 10-year graft rate and patient survival rate after LT for AALD are at least comparable to those of other indications. Nevertheless, long-term survival rates are hampered by frequent and/or excessive relapse in alcohol consumption. Relapse increases the risk of recurrent alcohol-associated cirrhosis but also of de novo alcohol-induced solid malignancies, mainly cancers of the upper aero digestive tract. Graft and patient survival rates, especially long-term, are thus hindered by the occurrence of excessive relapse.

Relapse rates vary immensely between studies and there is a lack of standardization in the definition of its severity, mainly because it is impossible to define the boundaries/thresholds for "safe consumption". However, there is consensus that harm appears for alcohol intake exceeding three portions per day for males and two for females, for at least 100 days with a sense of loss of control. This pattern of relapse, often described as "severe" can be found between 10 and 26% of patients. Most efforts aiming to reduce post-LT relapse rates focus on improving patient selection. Risk-factors of alcohol relapse often found in literature include short duration of pre-LT sobriety (<6 months), diagnosis of alcohol dependence, family history of alcohol-use disorder, psychiatric comorbidities including other substance abuse, prior alcohol rehabilitation and female gender. Scores such as HRAR (High Risk Alcoholism Relapse) have also attempted to stratify relapse risk based on pre-LT risk factors. Unfortunately, these criteria are not sensitive enough and most patients who finally benefit from the intervention are in the low to medium risk groups.

It is therefore a priority to utilize also resources in the post-LT setting to decrease alcohol relapse since it is a frequent and relatively difficult to predict event, with a high impact on outcomes after LT. We hypothesize that post-transplant addiction specialist interventions in liver transplant patients with AALD as primary, secondary or tertiary indication will result in decreased regular and/or severe alcohol relapse rate two years post-LT. By extension, this could result in higher graft and patient survival rates, especially in long term.

More recently, our group has performed a retrospective analysis of three centers with different addiction follow-up practices suggesting a benefit on severe relapse rates of addiction specialist intervention after LT for AALD. However, the main limit of this work is the retrospective design with different follow-up periods and duration. We designed a multicenter superiority randomized controlled trial with 2 parallel arms:

• Interventional arm where participants are offered targeted addictology follow-up and participate in addiction consultations
• Control arm where participants have classical follow-up by the transplant specialists of the LT center during the post-transplant follow-up period

The randomization will be elaborated using 1:1 ratio and minimization method. It will be stratified on centers and alcohol consumption history.

According to a French cohort study of patients with liver transplantation for alcohol-related disorders, 25% of them relapsed at 2 years.To account for mortality censoring during follow-up, we will apply a 5% increase of the sample size to reach 720 participants (360 in each arm). The comparison of the primary endpoint between arms will be carried out using intention to treat principle. The time to relapse will be expressed using Kaplan-Meier curves in each arm, and compared using a log-rank test. The effect size will be estimated using Cox proportional.