Deficient expression of glycosylphosphatidyl inositol (GPI)-anchored proteins in aplastic anaemia (AA) patients has previously been reported to be associated with a poor response to immunosuppressive (IS) therapy. Here we report the response to IS therapy of 111 patients with AA and correlate this with GPI-anchored protein expression on peripheral blood cells by flow cytometry. A GPI-anchored protein deficient population was identified in 15% (17/111) of patients with AA who had a negative Ham's test and no laboratory evidence of haemolysis. Patients were treated with antilymphocyte globulin and/or cyclosporin A, or oxymethalone. Bone marrow transplantation was performed in 12 patients, seven of whom had not responded to IS therapy. In patients tested for GPI-anchored protein expression prior to IS therapy there was no difference in response rate to IS therapy between AA patients with a GPI-anchored protein deficiency and those with normal GPI-anchored protein expression (50% response rate versus 75%, respectively). Survival in these two groups was similar at 90% with follow-up over 140 months from diagnosis. Eight of the 17 AA patients who developed a GPI-anchored protein-deficient population later went on to develop a positive Ham's test. From this study we demonstrate a lower incidence of GPI-anchored protein deficiency in AA patients compared with previous reports. In addition we have shown that the presence of a GPI-anchored protein-deficient cell population in patients with AA who have a negative Ham's test is not a poor prognostic factor in terms of response and survival after IS therapy.