Recombinant human granulocyte-macrophage colony-stimulating factor ameliorates zidovudine-induced neutropenia in patients with acquired immunodeficiency syndrome (AIDS)/AIDS-related complex.

To evaluate the effect of recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF) on patients with acquired immunodeficiency syndrome (AIDS) or AIDS-related complex (ARC) who were intolerant to zidovudine because of neutropenia, we performed a randomized, open-label study in which patients were assigned to one of two groups. Zidovudine was discontinued in group A patients before instituting GM-CSF treatment and was restarted in a graduated fashion over 4 weeks. Group B patients continued on full-dose (1,200 mg/d) zidovudine therapy while beginning GM-CSF therapy. A total of 17 patients were entered, eight in group A and nine in group B. Five of eight patients in group A and seven of nine in group B had a history of Pneumocystis carinii pneumonia (PCP). All were homosexual males, except one female in group A who was the sex partner of a bisexual male with AIDS. All patients had neutropenia (absolute neutrophil count [ANC] less than 1,000/microL) while taking full-dose zidovudine. The mean CD4 (+/- SD) lymphocyte level was 37 (+/- 29)/microL and 39 (+/- 44)/microL in groups A and B, respectively. After randomization, patients were begun on subcutaneous GM-CSF at a dose of 1.0 microgram/kg/d. Patients in group A received 2 weeks of daily GM-CSF, at which time zidovudine was restarted if the ANC was greater than 1,000/microL; if the ANC was less than 1,000/microL, the dose of GM-CSF was increased to 3.0 micrograms/kg, and at 2-week intervals either zidovudine was restarted or the dose of GM-CSF was increased to 5 micrograms/kg and then 10 micrograms/kg, to maintain the ANC greater than 1,000/microL. Group B patients received full-dose zidovudine concurrently with GM-CSF administration. The dose of GM-CSF was increased every 2 weeks if necessary to keep the ANC greater than 1,000/microL while maintaining full-dose zidovudine therapy. Patients in each group showed an increase in total white blood cell (WBC) count. Neutrophils and eosinophils were responsible for the majority of this increase. Patients in group A had a more rapid increase in WBC than those in group B; however, by week 8, the WBC in each group was essentially equal. Viral replication as measured by human immunodeficiency virus (HIV) p24 antigen (Ag) was decreased in four patients in each group, increased in one patient in each group, and remained unchanged in the remainder. The ability to culture virus from peripheral blood mononuclear cells was not changed by the regimen. The major toxicities of the regimen were fever and malaise.(ABSTRACT TRUNCATED AT 400 WORDS)