In the globe, about 33% (2 billion) of population has ever been infected with hepatitis B virus (HBV), and about 5% (350-400 million) were chronical HBV infection. In areas with high prevalence of hepatitis B, up to 80% of primary liver cancers are associated with HBV infection. About 25% of chronic hepatitis B virus carrier (more than 1 million people per year) eventually die of end stage liver disease associated with HBV infection, such as liver failure associated with cirrhosis and hepatocellular carcinoma. HBV replicates in the liver, which increases the risk of hepatocellular carcinoma in HBV carriers. Studies have shown that the risk of hepatocellular carcinoma (HCC) in HBV carriers was 10-100 folds higher than that of non-carriers.

Clinically, there are primarily two types of antiviral drugs: α-interferons (plain and pegylated ([PEG-IFN]α-2a or α-2b) interferons) and nucleos(t)ide analogues (NUC) including lamivudine (LAM), adefovir dipivoxil (ADV), entecavir (ETV), telbivudine (LDT), tenofovir disoproxil fumarate(TDF) and tenofovir alafenamide fumarate(TAF). With the development and application of antiviral drugs in recent years, the basic goal of maintain suppression against virus replication has been achieved, and HBsAg loss is considered as function cure of antiviral therapy. However, data from clinical studies showed a very low cure rate of current antiviral drugs and a natural HBsAg loss usually is less than 3%. The vast majority of clinical patients require long-term antiviral treatment and have difficulties in treatment stop.

The AI data mining system innovated by the Holy Haid owns a ten-million-scaled database and utilizes dozens of HBV-associated targets to identify 100 drugs that are most closely to the targets among the 500 commercially available drugs. With the identified 100 drugs, Holy Haid (Ying-ying Li) and Beijing Tsinghua Changgung Hospital (Lai Wei) conducted a cytological verification in mice, which indicated that the HD042 (Celecoxib) at 20uM concentration can inhibit HBV DNA, HBsAg and HBeAg by 70.87%, 88.52% and 87.55% respectively, without significant cytotoxicity. Based on this, Beijing Tsinghua Changgung Hospital (Lai Wei) retrospectively analyzed 1,114,661 patients admitted to 304 hospitals in 107 cities of 21 provinces and municipalities from January 1, 2019 to October 31, 2020 and identified 19,692 patients with the results of two HBsAg tests available and an interval of over 30 days. Among these, 3,359 patients had ever took HD042 (Celecoxib). Further analysis showed that these 3,359 patients, and screened out 383 patients who were diagnosed of hepatitis B and excluded from tumor with two HBsAg levels > 0.05IU/ml but ≤1500IU/ml. Among these, 110 patients were prescribed for more than 5 Celecoxib doses (about 30 days of treatment). Among the 110 patients, we screened out 27 patients on Celecoxib for 12 weeks whose HBsAg expression decreased by 59.2% after 12 weeks, including HBsAg clearance rate (i.e., HBsAg decreased to < 0.05IU/ mL) up to 18.5%.

Celecoxib, a specific inhibitor of Cyclooxygenase 2 (COX-2), has been widely used in clinical practice as an anti-inflammatory and analgesic drug. Studies have shown that Celecoxib improves NASH by inhibiting inflammatory responses. In addition, some studies have also shown that COX-2 is highly expressed in hepatitis B related hepatocellular carcinoma, resulting in cancerous tissue microangiogenesis. Cytological test found that Celecoxib, as a COX-2 specific inhibitor, can inhibit the growth of liver cancer cells by induced apoptosis and cell cycle inhibition, and have a even stronger effect on HBsAg positive liver cancer cells. However, the inhibitory effect of Celecoxib on the hepatitis B surface antigen in patients with chronic hepatitis B remained controversial. Therefore, this study is designed to investigate the safety and efficacy of Celecoxib in the hepatitis B surface antigen loss and reduction in nucleoside-treated patients with chronic hepatitis B.

Inclusion Criteria:

1. Aged 18-65;
2. Males or females;
3. Clinically diagnosed as chronic hepatitis B before taking nucleos(t)ide analogues (Entecavir, Tenofovir Disoproxil Fumarate, Tenofovir Alafenamide Fumarate) (HBsAg and/or positive HBV DNA for over 6 months, consistent or recurrent ALT elevation, histology confirmed chronic hepatitis B);
4. AST and ALT≤10 x ULN;
5. Total bilirubin ≤2 x ULN;
6. Having been treated with nucleos(t)ide analogues (Entecavir, Tenofovir Disoproxil Fumarate, Tenofovir Alafenamide Fumarate) for more than 1 year;
7. 100IU/ml < HBsAg < 1500IU/ml;
8. HBV DNA < 20IU/ml;
9. Child-Pugh class A;
10. Willing to sign an informed consent form.

Exclusion Criteria:

1. Patients with known allergy to Celecoxib or Sulfonamide;
2. Patients with oral aspirin or other NSAIDS (non-steroidal anti-inflammatory drugs) induced asthma, urticaria or anaphylactic reactions;
3. Patients treated for perioperative pains post coronary artery bypass graft (CABG);
4. Patients with active gastrointestinal ulcer/hemorrhage;
5. Patients with severe heart failure;
6. Patients with myocardial infarction within 3 months prior to enrollment;
7. ALT >10 x ULN or total bilirubin >2 x ULN;
8. Patients with peripheral leukocyte and/or platelet counts lower than lower limits of normal (LLN);
9. Patients with severe diseases of visceral organs (included but not limited cardiovascular, lung, kidney, brain) and fundus lesions;
10. Patients with concurrent autoimmune diseases, psychosis, diabetes, thyroid dysfunction (hyperactivity or hypothyroidism);
11. Patients with definite or suspected liver cancer or other malignancies;
12. Patients with historically organ transplant or ready to undergo organ transplant;
13. Patients on immunosuppressants;
14. Female patients who are pregnant or intended to become pregnant within 2 years;
15. Patients with history of drug or alcohol abuse;
16. Child-Pugh class B or C (current or prior onset);
17. Patients with concurrent HIV infection;
18. Patients with other liver diseases (including but not limited to positive Hepatitis C antibody);
19. Patients who are unable or unwilling to provide informed consent form or comply with study requirement.

Individual participant data might be available upon request after approval of IRB