Re-evaluation of alpha-interferon treatment of chronic hepatitis B using polymerase chain reaction.

To re-evaluate the efficacy of 3 treatment trials involving 272 Chinese patients with chronic hepatitis B virus infection, serial serum samples were tested from 60 patients treated with alpha-interferon with or without prednisone priming and 12 control patients, who were negative for hepatitis B virus deoxyribonucleic acid using dot-blot hybridization. Serial samples were tested using nested polymerase chain reaction with primer sets chosen from the surface and core antigen coding regions. The 19 patients who did not show persistent serological change remained hepatitis B virus deoxyribonucleic acid positive using the polymerase chain reaction assay. Three of the 4 patients (75%) who lost hepatitis B surface antigen and 9 of 51 (17.6%) who lost hepatitis B e antigen became negative from 0 to 60 months after the e-seroconversion. All patients negative for the polymerase chain reaction assay had normal transaminase levels. Pooling the 3 trials together, 11 of 188 (5.9%) treated patients and 1 of 84 (1.2%) control patients became hepatitis B virus deoxyribonucleic acid negative. The difference was not statistically significant. As assayed by the polymerase chain reaction assay, patients who were treated with alpha-interferon with or without steroid priming and lost hepatitis B e antigen within 12 months were more likely to subsequently lose the virus completely from the serum (11 of 33) than those who lost hepatitis B e antigen after 12 months (none of 13; p = 0.029).