Serum p24 antigen levels in untreated and zidovudine-treated HIV-1 infected subjects.

HIV-1 p24 antigen levels were determined in sequential serum samples from 71 subjects with persistent p24 antigenaemia, obtained over a period of 6-36 months during which no antiretroviral treatment was given. Fifty-six subjects were initially symptom free: 39 remained so during follow-up and 17 developed AIDS. Fifteen subjects were studied from the moment AIDS was diagnosed. Median serum p24 levels rose in all groups. In 50/71 individuals, p24 levels rose with time; in 21 they declined. The mean change in p24 levels was +194%. Twelve of the symptom-free subjects were subsequently treated with 1000 mg zidovudine daily for 12 months, over which period p24 levels declined in all 12 (mean change -56%). The difference in number of subjects with rising and declining p24 antigen levels between the untreated and zidovudine treated group was significant (P less than 0.001). Triton X-100 pretreatment (which presumably enables detection of virion-associated p24 in addition to free p24) of serum samples from 2 subjects who seroconverted for p24 antigen, from 2 AIDS patients who did not receive antiretroviral treatment and from 2 zidovudine-treated symptom-free subjects, did not lead to a substantial increase in p24 levels, except in the case of one of the p24 antigen seroconverters. It is concluded that the decline in serum p24 levels reported to occur during treatment with putative antiretroviral drugs can largely be ascribed to this treatment.