FALSE-POSITIVE HEPATITIS-B SURFACE-ANTIGEN SCREENING LEST RESULTS IN PATIENTS RECEIVING GRANULOCYTE-COLONY-STIMULATING FACTOR

Background: Granulocyte-colon-stimulating factor (G-CSF) is used for t he mobilization of progenitor cells and granulocytes. False-positive h epatitis 8 surface antigen (HBsAg) enzyme-linked immunosorbent assays (ELISAs) (NML) from one manufacturer In individuals receiving G-CSF ha ve been observed. Study Design and Methods: Sixty-six autologous perip heral blood progenitor cell donors from 1994 were retrospectively revi ewed. Donors typically received 5 to 10 mu g of G-CSF per kg subcutane ously for 5 days before collection. Additional ELISA dilutional studie s (1-in-10, 1-in-100, 1-in-1000) with known HBsAg-negative serum were made with G-CSF. Testing was performed by the University of North Caro lina, the American Red Cross in Charlotte, NC, or the National America n Red Cross, Washington, DC. Results: Of the 66 patients, none reacted for antibody to hepatitis B care antigen, and 30 (45%) had a positive reaction in the ELISA. Surface antigen positivity was ''confirmed'' o n 6 of the 30 patients by neutralizing ELISA reactivity with an antibo dy to HBsAg test from the same manufacturer. In all cases, the clinica l presentation was not suggestive of hepatitis, and these individuals were not at high risk for hepatitis B. Twenty-seven of the 30 cases we re tested with a monoclonal HBsAg ELISA (AUSZYME) from another manufac turer in the peridonation period and did not react. In 1994, 256 autol ogous whole-blood donors not receiving G-CSF were similarly tested and only 1 (0.4%) had a positive reaction with the second manufacturer's HBsAg ELISA (p<0.001). Of this group, 41 patients with histories of ma lignancy were identified, which is comparable to the history of the pe ripheral blood progenitor cell donors in this study, and none of these blood donors tested positive for HBsAg (p<0.001). Dilutional studies with G-CSF produced no reactions. Conclusion: The NML HBsAg ELISA stud ied has an unacceptably high false-positive rate in patients or donors receiving G-CSF. The false reactivity of this assay appears to be an indirect consequence of G-CSF administration, which can also lead to s purious confirmation by the HBsAg neutralization assay from the same m anufacturer.