Elevated Alpha-Fetoprotein Levels in Children with Metabolic Dysfunction-Associated Liver Disease

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD, previously referred to as nonalcoholic fatty liver disease—NAFLD) is a chronic liver disease closely associated with increased adiposity. ¹ MASLD can lead to end-stage liver disease (ESLD) and hepatocellular carcinoma (HCC) in a small proportion of patients. However, as MASLD is highly prevalent and estimated to affect 25% of the world’s population, the actual number of patients with ESLD and HCC is considerable and rising.^2,3 A recent analysis predicted a 137% increase in HCC and 168% increase in decompensated cirrhosis due to MASLD between 2015 and 2030, ultimately rendering MASLD the fastest rising indication for liver transplantation in adults.^2–4 While ESLD and HCC are far less common in the context of pediatric MASLD, they do occur.^5,6 The exact characteristics of patients developing such advanced liver disease remain to be determined, but risk factors include severe obesity, hypopituitarism, excessive calorie and processed food intake, and sedentary lifestyle. ⁷ A recent meta-analysis found an increased incidence of HCC in patients with MASLD and advanced fibrosis/cirrhosis, compared with patients with MASLD without advanced fibrosis/cirrhosis, where advanced liver fibrosis refers to stage 3–4.^8,9 The incidence of HCC was also higher in patients with MASLD with comorbid type 2 diabetes mellitus, especially when advanced fibrosis/cirrhosis was present. ⁸ As such, screening for progression to ESLD/HCC is pursued, particularly in high-risk individuals.

Alpha-fetoprotein (AFP) is a commonly used biomarker to screen for HCC. AFP is a glycoprotein with several important functions and implications across the lifespan. Elevations in AFP in childhood may indicate various pathologies, such as ESLD and hepatic malignancies (e.g., HCC, hepatoblastoma, and intrahepatic cholangiocarcinoma), as well as germ cell tumors (e.g., testicular cancers). ¹⁰ Considering that MASLD can lead to cirrhosis and HCC, and that adolescence (when children are typically screened for MASLD^1,11) is also a time when conditions such as germ cell tumors arise, an elevated AFP should be investigated further. Little is currently known about the prevalence of elevated AFP levels in youth with MASLD, particularly those with severe/advanced disease, and what the outcomes of these patients are.

The objectives of this study are to (1) investigate the prevalence of elevated AFP in children with advanced fibrosis in the context of MASLD, and (2) ascertain whether pediatric MASLD is associated with AFP elevations in the absence of advanced disease. We hypothesize that elevated AFP not only occurs in patients with advanced fibrosis but also occurs across the pediatric MASLD spectrum.

Materials and Methods

This was a retrospective cohort study performed at the Steatohepatitis Center of Cincinnati Children’s Hospital Medical Center (CCHMC). All research activities followed the approval of the protocol by the institutional review board (IRB #: 2023–0066).

Patients were included if they received care at the Steatohepatitis Center at CCHMC between January 2000 and January 2024, were 18 years of age or younger at the time of establishing care, and had at least one AFP level measured. To meet objective #1, patients were included if they underwent liver biopsy and were found to have fibrosis stage 3 or 4. To meet objective #2, patients were included if they had at least one elevated AFP level at any point during their care at the Steatohepatitis Center. Exclusion criteria were additional liver diseases or known malignancies at the time of liver biopsy or AFP measurement.

The electronic medical record was reviewed to collect demographic (age, ethnicity, sex), clinical (anthropometrics including body mass index [BMI]), laboratory (all available serum AFP levels, as well as levels of serum aminotransferases, fasting glucose, insulin and lipid panel, and HbA1c within 3 months of the AFP measurements), and concurrent histology data (when available). Additional data were collected in those with any elevated AFP level throughout the course of their clinical follow-up at CCHMC, including all follow-up AFP levels, as well as any available, concurrent imaging studies such as ultrasonography, computed tomography (CT), or magnetic resonance imaging (MRI).

At our institution, AFP levels are measured in all patients with suspected MASLD at baseline and are repeated per the clinicians’ discretion (e.g., in patients with advanced liver disease). Per our institution’s laboratory, elevated AFP is defined as >8.0 ng/mL.

Results

One thousand three hundred eighty-nine patients aged 6–18 years followed in the Steatohepatitis Center for suspected MASLD were identified and screened for inclusion. Of those, 483 (34.8%) had AFP levels available. This cohort was 68% male and 65% non-Hispanic, with a median age of 14 years (interquartile range [IQR] = 11–16) at time of initial AFP level. Characteristics of the 483 patients with MASLD and AFP levels are summarized in Table 1. Of this cohort, 161 had undergone a liver biopsy at least once, and of those, 100 had fibrosis (stages 1–4). Twenty-two of those patients had advanced fibrosis (3–4).

Characteristics of the 22 patients with MASLD and advanced fibrosis are summarized in Table 1. Their median age was 11 years (IQR = 10–18), and they were predominantly male (n = 18; 82%) and non-Hispanic (n = 12; 55%). All had severe obesity with a median BMI z-score of 2.56 (IQR = 2.33–2.75) and mean NAFLD activity score of 5.2 (SD ± 1.4) at the time of biopsy. Most of those with advanced fibrosis had an AFP level measured after the liver biopsy (n = 14 of 22; 64%). AFP levels were obtained once (IQR = 1–3) at median 19 months (IQR = 10–45 months) after biopsy. Seven patients had an AFP level checked most recently at median 8 months (IQR = 5–13 months) prior to liver biopsy, and one patient had an AFP level checked at the time of biopsy. None had elevated AFP at any time, and none were diagnosed with malignancy.

Of the entire cohort of patients with MASLD (n = 461, excluding those with advanced fibrosis characterized above), nine were found to have an elevated AFP level at least once during their follow-up. They had a median age of 15 years (IQR = 13–18) at the time of initial AFP elevation and were predominantly male (78%) and non-Hispanic (67%). All had severe obesity with a median BMI z-score of 2.56 (IQR = 2.07–2.71). Two patients had diabetes at the time of AFP elevation, and the rest had an HbA1c within the normal range. Median initial elevated AFP level was 10.8 ng/mL (IQR = 9.9–13.5), and peak AFP level was 11.1 ng/mL (IQR = 10.4–14.3). Of those, four patients underwent liver biopsy, and only one was found to have fibrosis (stage 1). Work-up and outcomes are summarized in Table 2.

Discussion

In this study, we investigated the use of AFP as a screen for HCC in pediatric patients with MASLD followed at the Steatohepatitis Center of CCHMC. First, we looked at AFP levels in those with advanced fibrosis. Of this predominantly adolescent, male, and non-Hispanic cohort with severe obesity, none had an elevated AFP, and none were diagnosed with malignancy. Monitoring of liver disease progression and HCC surveillance varied in this cohort; however, at a median follow-up of 19 months, all patients tested continued to have normal AFP levels. In contrast, AFP was elevated in a small proportion of the general MASLD population studied. The elevations were mild, and none were associated with the presence of malignancy. The work-up to rule out the latter was also variable.

AFP was first discovered to be a marker of primary liver tumors in the 1960s and has since been widely used to diagnose, assess treatment response, and monitor surveillance of HCC and hepatoblastoma.^12,13 AFP is initially produced by the yolk sac during fetal development, until the fourth week of gestation when the fetal liver becomes the primary site of synthesis.^12,14 In the fetus, AFP serves as the primary serum-binding protein and transports hormones, fatty acids, and other molecules to aid in cell growth and development. ¹² When present in adults, these functions promote cell proliferation and motility, thereby supporting the theory that AFP is pro-oncogenic. ¹³ After birth, AFP is replaced by albumin as the primary serum-binding protein, and AFP levels drop to adult levels after 2 years of age.^10,12 As AFP is produced by primary hepatic tumors, as well as yolk sac and germ cell tumors, elevated levels in older children and adults raise concern for malignancy. ¹² Thus, measuring AFP is routine in patients with risk factors for hepatic malignancies.

Although it is the only routinely recommended biomarker for HCC, AFP is an imperfect screening tool. ¹⁵ As a standalone test for early-stage HCC, at the cutoff of 20 ng/mL, AFP’s sensitivity ranges from 39% to 64% and specificity from 76% to 97%. ¹⁶ At higher cutoff values (e.g., 400 ng/mL), specificity reaches 99%; however, sensitivity declines to 17%. ¹³ Additionally, a meta-analysis has shown that AFP increases abdominal ultrasound’s sensitivity for HCC screening incrementally, from 45% to 63%.^16,17 Furthermore, there is lack of standardization in the cutoff AFP value that is considered clinically significant. AFP’s poor sensitivity and specificity are also highlighted by studies in adults showing that AFP levels are higher in those with MASLD and metabolic syndrome compared with healthy controls.^18–20 In that context, AFP elevation may be secondary to chronic inflammation and tissue regeneration. ¹⁸ Extensive or chronic inflammation has been linked to AFP elevations.^19,21 Whether the latter is the reason why our patients with cholecystitis and biliary ductal dilation presented with AFP elevation remains to be shown. ²¹ There is currently no evidence of association between epididymal cysts and elevated AFP. Overall, these data underscore the limitations of AFP as a screening tool for HCC in adolescents with MASLD-related fibrosis. AFP elevations in this context should be interpreted with caution and not immediately raise concern.

MASLD has become a leading cause of ESLD requiring liver transplantation in adults.^5,7,22 Widely accepted recommendations for HCC surveillance in patients with ESLD are to follow AFP levels and obtain abdominal imaging (ultrasound, MRI, or CT) every 6–12 months.^23,24 While uncommon, there are reports of children developing cirrhosis and HCC; however, the risk factors for this progression have not been entirely elucidated.^5,6,23 As such, there is no clear guidance with regards to if, when, and how often clinicians should screen for these complications. Our results suggest that the risk of HCC is exceedingly low in pediatric MASLD cohorts, even in the context of advanced fibrosis. We do note however that, as expected, our sample size of patients with advanced fibrosis was small, limiting our ability to make any formal recommendations with regards to screening. Given that ESLD and HCC are more likely to develop in those with specific underlying conditions (e.g., panhypopituitarism/growth hormone deficiency) or more advanced disease (e.g., clinically significant fibrosis of stage 2 or higher), it is reasonable to limit screening to those subgroups.^5,7,25

Strengths of this study include the large cohort of patients with MASLD and the decreased risk of selection bias, as all patients who had an AFP level drawn were included. Limitations include the retrospective nature of our study, relatively few AFP levels checked after histological diagnosis of advanced fibrosis, as well as the inability to study the patients who were lost to follow-up and might have received work-up elsewhere.

In conclusion, in our cohort, AFP was not elevated in patients with MASLD and advanced fibrosis. Mild but clinically not significant elevations of AFP can be seen with pediatric MASLD, however. Our findings and the rarity of primary liver tumors in children, adolescents, and young adults with MASLD suggest that the clinical focus should remain on monitoring progression to advanced fibrosis and ESLD rather than underlying malignancy.

Impact Statement

Obesity is a risk factor for developing advanced liver disease, including end-stage liver disease and hepatocellular carcinoma, in the setting of metabolic dysfunction-associated steatotic liver disease (MASLD). As the prevalence of MASLD rises, the need for hepatocellular carcinoma screening guidelines in adolescents and young adults becomes crucial.