Clinical features of abnormal α-fetoprotein in 15 patients with chronic viral hepatitis B after treatment with antiviral drugs

Abstract

BACKGROUND

Liver function of chronic hepatitis B (CHB) patients is essentially normal after treatment with antiviral drugs. In rare cases, persistently abnormally elevated α-fetoprotein (AFP) is seen in CHB patients following long-term antiviral treatment. However, in the absence of imaging evidence of liver cancer, a reasonable explanation for this phenomenon is still lacking.

AIM

To explore the causes of abnormal AFP in patients with CHB who were not diagnosed with liver cancer.

METHODS

From November 2019 to May 2023, 15 patients with CHB after antiviral treatment and elevated AFP were selected. Clinical data and quality indicators related to laboratory testing, imaging data, and pathological data were obtained through inpatient medical records.

RESULTS

All patients had increased AFP and significantly elevated IgG. Cancer was excluded by imaging examination. Only four patients had elevated alanine aminotransferase, 10 had elevated aspartate aminotransferase, nine had elevated total bilirubin, and two had antinuclear antibodies. The liver biopsy and histopathological examination indicated that 14 patients had rosette, moderate, or higher interfacial inflammation, lymphocyte infiltration, and severe hepatic fibers (11 cases), which was consistent with the pathological features of autoimmune hepatitis (AIH). After 8-12 week of hormone therapy, the levels of AFP and IgG, and liver function returned to normal (P < 0.05).

CONCLUSION

For patients with CHB and elevated AFP after antiviral treatment, autoimmune hepatitis should be considered. CHB with AIH is clinically insidious and difficult to detect, and prone to progression to cirrhosis. Liver puncture pathological examination should be performed when necessary to confirm diagnosis.

Key Words: Abnormal α-fetoprotein; Chronic viral hepatitis B; Antiviral treatment; Autoimmune hepatitis; Hepatic pathology

Core Tip: There is no reasonable explanation for the continuous abnormal increase of α-fetoprotein (AFP) in patients with chronic hepatitis B (CHB) after long-term antiviral therapy and the lack of imaging evidence of liver tumor. We observed 15 patients with CHB treated with antiviral therapy without imaging evidence of liver cancer, and explored the cause of elevated AFP by liver biopsy. Pathological examination suggested that autoimmune hepatitis (AIH) might be complicated, but the diagnosis could not be confirmed by a comprehensive and a simplified diagnostic scoring system for AIH.

INTRODUCTION

Chronic hepatitis B (CHB) is the largest public health burden in China. Long-term treatment with antiviral drugs inhibits HBV DNA replication and controls liver inflammation in most CHB patients. The indexes reflecting liver damage, such as alanine aminotransferase (ALT), aspartate aminotransferase (AST), and α-fetoprotein (AFP) decrease and return to normal levels[1]. However, in rare cases, a sustained abnormal increase in AFP is detected in patients with CHB following long-term antiviral treatment, while there is a lack of imaging evidence of liver tumors, and thus no reasonable explanation for this phenomenon exists. In this study, we of clinical examination results and liver biopsy evidence to further explore this problem.

MATERIALS AND METHODS

Study population

We enrolled 15 CHB patients with elevated AFP who underwent liver biopsy at the Fifth People's Hospital of Ganzhou City between November 2019 and May 2023. CHB is defined as positive hepatitis B surface antigen (HBsAg) for > 6 months[2]. Patients received antiviral therapy with oral nucleoside analogs. Exclusion criteria were: history of excessive alcohol consumption (defined as ≥ 20 g/d for men and ≥ 10 g/d for women, or 2 weeks of heavy drinking ≥ 80 g/d); drug-induced hepatitis; schistosomiasis liver disease; liver cancer; Human immunodeficiency virus infection or hepatitis A, C, or D; pregnancy; as well as patients receiving corticosteroids, chemotherapy or immunosuppressive therapy, liver transplantation, or hemodialysis. This study was approved by the Ethics Committee of the Fifth People's Hospital of Ganzhou City.

Data collection

The following clinical data were collected: Age; sex; antiviral treatment; and results of routine blood tests and serological examination. Liver function indicators were collected before hormone therapy, and 1 and 2-3 months after treatment, including total bilirubin (TBIL), albumin (ALB), ALT, AST, total bile acid (TBA), γ-glutamyltransferase (GGT), alkaline phosphatase (ALP), AFP, other viral hepatitis markers, autoantibodies (antinuclear antibody), and IgG. Imaging, such as computed tomography or magnetic resonance imaging, was performed.

Integral diagnosis of autoimmune hepatitis

A comprehensive diagnostic scoring system (1999)[3] and a simplified diagnostic scoring system (2008)[4] were used to diagnose autoimmune hepatitis (AIH).

Pathological assessment

All assessments were carried out by specialists who had extensive experience in ultrasound diagnosis. A 16G puncture needle (Bard Peripheral Vascular Inc., Tempe, AZ, United States) was used for percutaneous liver biopsy. The length of the obtained tissue exceeded 1.5 cm. After fixation with formalin, the tissue underwent standard procedures including dehydration, paraffin embedding, sectioning, as well as routine hematoxylin and eosin, Masson, and reticular fiber staining. Subsequently, two liver pathologists evaluated the tissue based on the Ishak Histological Scoring System.

Statistical analysis

Measurement data conforming to normal distribution were presented by mean and standard deviation. Comparison between the two groups was performed by Student's t-test. If the distribution did not conform to normal, the Mann-Whitney test was used to compare continuous variables. The χ² test was used to compare categorical variables. P < 0.05 was considered statistically significant.

RESULTS

Clinical features

We enrolled 15 patients diagnosed with CHB, 10 men and five women, with an average age of 54 years (range 48-69 years). Four patients showed elevated ALT levels, 10 elevated AST, nine elevated TBIL, six elevated ALP, and 10 elevated GGT. All patients underwent antiviral treatment prior to liver biopsy, and eight achieved a complete HBV response (HBV-DNA < 10 U/L), five had HBV-DNA levels < 2 × 10³ U/L, and two had HBV-DNA levels > 2 × 10³ U/L. Seven patients with CHB received initial antiviral therapy (≤ 6 months), including five with entecavir, one with tenofovir, and one with tenofovir amibufenamide. Eight patients received long-term antiviral therapy (≥ 6 months), including seven who were managed with entecavir and one with tenofovir alafenamide fumarate. Two patients ANA positive (1:320 and 1:160) and 13 were autoantibody negative (Table 1).

Table 1 Baseline clinical data of 15 patients with chronic hepatitis B combined with elevated -fetoprotein.

No.	Sex	Age	ALT	AST	TBIL	ALP	GGT	TBA	ALB	AFP	ANA	IgG	HBV-DNA
1	Female	62	29.9	41.3	8.9	98.9	281.7	24.5	41.4	29.77	1: 320	18.73	30.8
2	Male	48	61.9	100.3	27	114.4	57.5	82.6	27.9	32.96	-	30	-1
3	Male	56	35.7	44	58.6	140.9	106.8	78.3	27.9	19.36	-	25.1	-1
4	Female	64	77	69.7	24.3	143.7	100.1	67.9	25.9	142.96	-	34.7	6.49 × 103
5	Male	57	34.5	50.3	46.5	92	93.3	299.5	29.8	85.75	-	26	-1
6	Female	57	28.5	65	44.5	149.9	185.8	45.6	25	77.42	1: 160	21.15	1.27 × 102
7	Male	67	15.5	21.8	18.8	71.8	48	34.8	34	8.47	<1: 80	21.8	-1
8	Male	57	22.7	57.2	20.1	107.7	46.2	39.7	33	42.99	<1: 80	28.03	5.48 × 102
9	Male	60	34.3	66	45.8	157.8	92	60.9	24.1	18.56	-	22.32	-1
10	Female	69	20.6	27	12.6	178.5	41	2	22.3	14.28	<1: 80	25.23	-1
11	Female	51	39.5	35	17.2	73.1	51	33.5	40.3	6.85	<1: 80	20.27	3.06 × 102
12	Male	49	24.7	30	33.1	111.1	134	3.8	28.8	340.1	<1: 80	19.89	51.5
13	Male	52	36.4	104	50.5	148.8	179	12.5	36.9	1004	<1: 80	24.48	5.24 × 103
14	Male	57	72.6	63	8.1	59.1	48	70.8	40.5	158.3	<1: 80	22.04	-1
15	Male	50	24	32	30.3	123.7	65	8	35.1	10.69	<1: 80	25.27	-1

¹Hepatitis B virus DNA level below lower limit of detection.

ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TBIL: Total bilirubin; ALP: Alkaline phosphatase; GGT: Gamma-glutamyltransferase; TBA: Total bile acid; ALB: Albumin; AFP: Alpha fetoprotein; ANA: Antinuclear antibody; HBV-DNA: Hepatitis B virus DNA.

Pathological characteristics

The pathological alterations of the liver were primarily attributed to necrotizing inflammation around the portal area or the surrounding septal region. All patients demonstrated predominant liver cell injury attributed to interfacial hepatitis. Fourteen patients presented with severe interfacial necrotizing inflammation along with lymphoplasmacytic cell infiltration. Fusion necrosis was observed in 14 patients, with 13 experiencing severe fusion necrosis (≥ 4 points). A typical rosette (florid duct lesion; FDL) was observed in 14 patients. All 15 patients exhibited pathological manifestations of lobulitis and portal inflammation, with 11 demonstrating grade 3/4 lobulitis and 13 showing grade 3/4 portal inflammation. None of the 15 patients had bile duct disappearance but four had bile duct injury and nine had cholestasis (Figure 1). The mean Ishak histological score was 14.38. According to the histopathological diagnosis of AIH, the liver puncture pathological morphology in 14 patients was consistent with that of AIH, while one patient exhibited atypical findings. Intrahepatic inflammation was graded at G3 and above in 14 (93.33%) patients. Fourteen patients had intrahepatic fibrosis grade above S3 (Table 2).

Figure 1 Histological characteristics of liver in patients 8, 9, 10, and 12 (a, b, c, and d). A: MUM-1 immunohistochemistry: A large number of plasma cells infiltrated in the hepatic lobules, and plasma cells concentrated in the sink area and interface (blue arrow); B: Hematoxylin and eosin staining: Rosettes (yellow arrow) extensive inflammatory cell infiltration in the hepatic lobules, moderate to severe interfacial hepatitis (black arrow).

Table 2 Pathological features of 15 patients with chronic hepatitis B combined with elevated alpha fetoprotein.

No.	II	Lym/PCs I	FN	BN	FDL	LI	PI	BDI	ABD	Fibrosis	Ishak score	G	S
1	3	Yes	2	0	1	4	3	0	0	3	12	G3	S1
2	3	Yes	6	1	1	2	3	0	0	6	15	G3-4	S4
3	4	Yes	5	1	1	2	3	0	0	6	15	G3-4	S4
4	4	Yes	5	1	1	3	4	0	0	5	17	G3-4	S3-4
5	4	Yes	5	1	1	3	3	0	0	5	16	G2-3	S4
6	4	Yes	5	1	1	4	4	0	0	5	18	G4	S3-4
7	4	No	4	1	1	2	4	0	0	6	15	G2-3	S4
8	4	Yes	6	1	1	3	3	0	0	6	17	G4	S4
9	4	Yes	4	1	1	3	3	1	0	6	15	G3-4	S4
10	3	Yes	4	1	1	4	3	1	0	3	15	G3-4	S3-4
11	3	Yes	6	1	1	3	4	1	0	6	17	G3	S4
12	4	Yes	4	1	1	3	4	1	0	6	16	G4	S4
13	4	Yes	5	1	1	3	4	0	0	6	17	G3	S3-4
14	3	Yes	1	0	1	2	2	0	0	3	8	G3	S3-4
15	2	Yes	0	0	0	1	2	0	0	3	5	G2	S3-4

II: Interfacial inflammation; Lym/PCs I: Infiltration of lymphoid/plasma cells; FN: Fusion necrosis; BN: Bridging necrosis; FDL: Florid duct lesion; LI: Lobular inflammation; PI: Portal inflammation; BDI: Bile ductal injure; ABD: Absence of bile duct.

AIH classification based on the diagnostic scores

AIH diagnostic scores were calculated for all 15 patients using both the AIH simplified diagnostic score system and the integrated diagnostic score system. In their respective scoring systems, two patients were classified as probable AIH using the simplified diagnostic score system, while four were classified as probable AIH using the integrated diagnostic score system. None of the patients were diagnosed as “clear” with either diagnostic scoring system (Table 3). The specificity of these diagnostic scoring systems could not be assessed as we only analyzed patients with coexisting AIH and CHB.

Table 3 Autoimmune hepatitis diagnostic scores of 15 patients.

No.	Integrated diagnostic score		Simplified diagnostic score
	Pre-treatment	Post-treatment
1	13	15	6
2	9	11	4
3	9	11	4
4	12	14	4
5	9	11	4
6	10	12	6
7	9	11	4
8	9	11	4
9	8	10	4
10	9	11	4
11	10	12	4
12	8	10	4
13	9	11	4
14	8	10	4
15	9	11	4

Treatment and outcome

After undergoing liver biopsy for the consideration of AIH diagnosis, 14 patients received treatment with immunosuppressants (prednisone 20-40 mg/d alone or 50 mg/d in combination with azathioprine). After 8-12 week, all patients showed improvement in symptoms. There was a significant decrease in AFP and IgG levels (Table 4), as well as decreased ALP and AST levels, and increased ALB levels (all P < 0.05) (Table 5).

Table 4 Changes in -fetoprotein before and after hormone therapy.

Parameters	Pre-hormone therapy	1 months after hormone therapy	2-3 months after hormone therapy	Z/t	P value
AFP (μg/L)	32.96 (12.49-114.36)	15.38 (11.21-41.71)	5.2 (3.94, 9.44)1	3.18	0.001

¹Using single-factor repeated measures one-way analysis of variance, the difference in alpha fetoprotein index between 2 and 3 months after hormone therapy and before treatment was significant.

AFP: Alpha fetoprotein.

Table 5 Biochemical index changes before and after hormone therapy, n (%)/mean ± SD.

Parameters	Pre-hormone therapy	2-3 months after hormone therapy	Z/t	P value
IgG	24.83 ± 4.32	16.09 ± 2.87	16.263	< 0.001
ALT	41 (33.85-255.5)	34.7 (28.5-39.5)	1.153	0.249
AST	51 (39-221)	33 (27.5-43)	2.274	0.023
TBIL	24.3 (14.9-45.5)	18.5 (13.15-46.25)	0.454	0.65
ALP	115.58 ± 35.99	82.72 ± 28.13	4.035	0.002
GGT	100.26 ± 72.99	63.22 ± 24.96	1.591	0.138
TBA	39.7 (18.5-74.55)	49.2 (12.95-90.15)	0.874	0.382
ALB	32.31 ± 6.36	36.18 ± 4.57	2.983	0.011

ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; TBIL: Total bilirubin; ALP: Alkaline phosphatase; GGT: Gamma-glutamyltransferase; TBA: Total bile acid; ALB: Albumin.

DISCUSSION

AFP is a glycoprotein that is synthesized by the liver and found in fetal serum. Abnormal elevation of AFP is often associated with liver injury. Liver injury can result in the activation and proliferation of liver precursor cells, leading to an increase in AFP secretion. Elevated levels of human AFP are significantly linked to germ cell tumors, such as yolk sac tumors[5] and testicular cancer[6]. Additionally, increased AFP is commonly observed in pregnant women. In several studies involving patients with liver cancer, it was found that AFP levels were significantly elevated. Subsequent research has demonstrated that AFP is a highly specific and sensitive marker for hepatocellular carcinoma[7]. A significant number of patients with acute and chronic hepatitis and cirrhosis without malignancy have elevated serum AFP levels, which subsequently decrease after the recovery of liver inflammation. Therefore, an elevated level of AFP is closely associated with liver injury, indicating increased DNA synthesis or the regeneration of liver cells.

In this study, all 15 patients received antiviral therapy, and imaging examination ruled out the possibility of a tumor, but AFP levels remained abnormal. Some reports suggest that the incidence of abnormally elevated serum AFP levels in patients with active viral hepatitis is significantly reduced after the initiation of antiviral therapy[1,8,9]. Qian et al[10] conducted a study on 3680 patients with non-HCC viral hepatitis from five central hospitals and found that antiviral treatment was associated with a decrease in AFP levels (2.68 ng/mL). Therefore, the 15 patients in our study had a decreased likelihood of elevated APF and active viral hepatitis. AFP levels may be elevated in patients with advanced fibrosis and cirrhosis, although it is typically a transient phenomenon[11,12]. In this study, 11 patients were found to have advanced liver fibrosis. However, the increased AFP levels persisted for several weeks, and the correlation between increased AFP and advanced liver fibrosis was deemed insufficient. During the examination of autoantibodies, a persistent increase in IgG levels was observed. Antiviral therapy led to a decrease in transaminase and viral load in 15 patients, but serum AFP, transaminases, and IgG levels were fully normalized only after immunosuppressive therapy. These findings are consistent with the study results of Efe et al[13]. In patients with active HBV combined with AIH, antiviral therapy alone could not completely restore ALT, AST, and IgG levels to normal; therefore, combined immunosuppressive therapy was necessary[13]. This suggests that biochemical indicators of liver inflammation may not fully reflect liver injury. Our research also indicates that elevated AFP can serve as a marker for liver inflammation when accompanied by mildly abnormal aminotransferase levels and no basis for diagnosing liver cancer.

The clinical diagnosis of AIH relies on the AIH comprehensive and simplified diagnostic scoring systems, with liver pathological features serving as key evidence. Both CHB and AIH are chronic inflammatory lesions mainly caused by liver cell injury[14]. The pathological features of CHB are characterized mainly by necrotizing and lobular inflammation in the portal vein area, which is nonspecific and challenging to differentiate from other chronic liver injuries. The primary histological characteristics of AIH include interface hepatitis, infiltration of lymphocytes/plasma cells, and FDL. Compared to other causes of interfacial hepatitis, AIH interfacial inflammation is typically more prominent and can be observed in up to 98% of AIH patients. This is considered the hallmark of AIH, and moderate to severe interfacial hepatitis often supports the diagnosis of AIH[15]. In this study, 14 of the 15 patients had moderate or higher interfacial inflammation, significant plasma cell/Lymphocyte infiltration in the portal area, and rosette-like changes in liver pathology, which were consistent with the characteristic liver pathological characteristics of AIH and supported the diagnosis of AIH. None of patients met the diagnostic criteria for AIH according to the International Autoimmune Hepatitis Group (IAIHG) (1999) and the simplified IAIHG (2008) score. The diagnosis of AIH was an excludable diagnosis. All 15 patients in this study had coexistent CHB, which was not suitable for the simplified scoring system in 2008. In these diagnostic criteria, the positive score of viral hepatitis was -2 points, and the maximum score was only 6 points. The traditional scoring system of the IAIHG in 1999 was used for the integral diagnosis. Positive viral hepatitis would score -3 and there were other negative components. Therefore, only four patients (26.67%) were diagnosed as possible AIH, and none were confirmed as AIH. However, the diagnosis of AIH in 15 patients was considered according to biochemistry, autoantibodies, IgG, and liver histology. Therefore, both scoring systems provide limited assistance in diagnosing AIH in patients with concomitant chronic viral hepatitis, thereby complicating the treatment decision-making process. Due to the consistent liver pathology of 15 patients with characteristic AIH features, glucocorticoid therapy was administered to 14 patients for a definitive diagnosis. After a follow-up period of 2-3 months, AFP and IgG levels gradually returned to normal, suggesting that the diagnosis of CHB combined with AIH should be considered in this study. Consequently, the diagnosis of viral hepatitis combined with AIH demands a comprehensive diagnosis, among which liver biopsy histopathology is particularly significant.

Patients with CHB combined with AIH exhibited more severe histological inflammation than in patients with either condition alone, often accompanied by active inflammation. Previous studies conducted abroad have indicated that hepatitis virus infection may lead to autoimmune liver disease and exacerbate its progression. Additionally, there is a higher likelihood of advanced liver fibrosis[16]. In another study of 139 cases of autoimmune liver disease, the liver pathology results showed that the proportion of patients with hepatitis virus infection and liver fibrosis stage F3/F4 was significantly higher than that of patients with pure autoimmune liver disease[17]. This suggests that the degree of liver inflammation in patients with CHB combined with AIH is often more severe than that in patients with single AIH or CHB. However, the proportion of patients with AIH and advanced liver fibrosis in Sbikina et al’s study was significantly lower than that in our study (93.33%), as the study encompassed HBsAg-negative patients with previous HBV infection[17]. There is a continuous inflammatory response to liver cells and liver cell necrosis, which leads to an increase in AFP production, promotes the immunoregulatory function of the body, and enhances the repair and regeneration ability of liver cells. According to a study by Ma Xiong et al[18] in China, it was found that 94.12% of patients with CHB combined with AIH had liver inflammatory activity above G3. This further indicates that patients with CHB combined with AIH exhibit relatively active liver inflammatory activity.

One drawback of this study was the small sample size. In addition, the follow-up time was short, and the effect of long-term immunotherapy on anti-HBV therapy in CHB patients with AIH was not observed. The number of AIH patients with CHB combined with increased AFP was not counted in this study, and the incidence of AIH in CHB combined with increased AFP was poorly understood. Therefore, further multicenter studies with sufficiently large sample sizes and long follow-up periods are needed to detect significant associations between the results.

CONCLUSION

This is the first case series to demonstrate an association between AFP and CHB with AIH. AIH based on CHB is prone to being overlooked. This study offers a novel diagnostic proposal for CHB with AIH. After treating CHB patients with persistently abnormal AFP and excluding tumor-related disorders, it is essential to contemplate the possibility of CHB combined with AIH and other types of chronic active hepatitis. If necessary, liver biopsy can be used to aid in diagnosis.