Malignant transformation of hepatocellular adenoma.

To the Editor:

We read with interest the paper by Chopinet et al. on long-term outcomes following resection of hepatocellular adenoma with malignant transformation (MT-HCA).

In their retrospective surgical study, resected specimens were divided into 2 groups: HCA with small foci of malignancy <1 cm (SF-HCA, 23 patients) and malignant HCA with macroscopic malignant nodules larger than 1 cm (M-HCA, 17 patients). Recurrence free-survival (RFS) after liver resection (LR) for MT-HCA was compared to patients resected for hepatocellular carcinoma (HCC) occurring on normal liver parenchyma (NP-HCC). Of these MT-HCA,16/40 (40%) had β-catenin mutations, 19/40 (47.5%) were inflammatory, 1 was HNF1A-mutated and 4 (10%) were unclassified HCA. Microvascular invasion (12% vs. 0%) and satellite nodules (25% vs. 4%) were more frequently observed in M-HCA than in SF-HCA. After a median follow-up of 67 months, 10 (25%) patients with MT-HCA had tumor recurrence, including 9 with M-HCA and 1 with SF-HCA. M-HCA was linked to significantly poorer RFS rates than SF-HCA. The authors identified age >55 years, presence of satellite nodes, and microvascular invasion as risk factors for long-term recurrence. This study has the major merit of being the only one available on long-term risk of recurrence in patients with MT-HCA following LR, compared to NP-HCC. However, as mentioned by the authors, this study has major bias.

The definitions of SF-HCA and MT-HCA were determined on operative specimens and defined by the extent of the malignant foci area. They ignore the number of foci, the size and number of malignant nodules; the notion of borderline lesions was not considered. Defining inflammatory HCA (IHCA) based on serum amyloid A positivity in at least 10% of tumorous hepatocytes will inevitably lead to the overestimation of its prevalence; shHCA were not classified.

The long study period between 2001-2019 is incompatible with the progress made on MRI, HCA classification, molecular analysis.

The low number of patients included (40 MT-HCA, and 40 NP-HCC) prevents any solid conclusion from being drawn.

This led us to review our own experience. Our HCA database includes 250 cases (33 men). All cases with the terms HCC and borderline lesions were retrieved. It includes only cases with obvious pathological proof that HCC was developed on HCA. These patients were regularly followed. Pathological (macroscopy and microcopy), surgical and clinical records were reviewed, as well as slides (H&E, reticulin, CD34, MIB1, immunomarkers for HCA classification.) Data on deaths related or not to HCC were collected as was follow-up data for patients alive without recurrence. Data are summarized in Table S1.

Contrary to Chopinet et al., we did not observe HCC in patients with IHCA, which is surprisingly the major subtype in Chopinet et al.’s paper. It is known that HCC, without any previous HCA, can frequently overexpress inflammatory proteins and exhibit low levels of LFABP, that can be misleading with the diagnosis of MT-HCA. The majority of our cases with malignant features were HCAs with a high level of β-catenin pathway activation (mainly due to exon 3 non S45 mutation or large deletion).

In our series, we observed a limited number of HCC (12 cases; 4.8%) and borderline HCA/HCC (17 cases; 6.8%). One could argue that our definition of borderline lesion could correspond to SF-HCA described by Chopinet et al. or identified by others as very well-differentiated HCC or HUMP.

In the majority of our MT-HCA, 20 to 80% of the tumoral volume were well differentiated with no microvascular invasion or satellite nodules.

Recurrence and death was explained in 2 cases by voluminous HCC compared to the total volume of the tumor (poorly differentiated HCC in one case), and by the lack of effective surgery in 2 cases. The fifth death was not related to HCC.

Data on long-term outcomes are important for the development of optimal management strategies for HCA, especially since derivation of NP-HCC from an HCA remains a challenge.[7], [8], [9], [10] Today, sex and size of HCA remain the most well-established parameters to guide treatment decisions. Elective liver resection is recommended in all men, while women are subjected to surgery when HCA diameter is greater than 5 cm and/or size has been progressing despite life-style changes. Nonetheless, multiple case reports showed malignant transformation in HCAs with diameters <5 cm, while other studies suggest an elective surgery for diameters >4 or >8 cm, opening a debate around the optimal cut-off values for surgical treatment. Selected patients obviously benefit from conservative therapy, such as the cessation of exogenous hormone intake.

The understanding of HCA risk factors has evolved gradually to include: demographic factors (age, sex,) environmental factors (oral contraception, drugs, anabolic steroids, obesity, metabolic syndrome, alcohol); underlying liver disease (non-alcoholic steatohepatitis, vascular liver diseases and even alcohol-related cirrhosis), genetic diseases (MODY3, glycogenosis, familial adenomatous polyposis…) and molecular characteristics. The time has come to individually assess the risk of hepatocarcinogenesis for each patient with HCA in order to provide risk-adjusted management and treatment. This can only be done at a European/international level.

Financial support

The authors received no financial support to produce this manuscript.

Author’s contributions

PBS and CB wrote the manuscript. PBS and BLB reviewed the pathological data. CJ, LC and CB collected the clinical data. All the authors approved the final document

Conflict of interest

The authors declare no conflicts of interest that pertain to this work.

Please refer to the accompanying ICMJE disclosure forms for further details.