HIV detection in culture of peripheral blood mononuclear cells: a relevant in vitro correlate of the antiviral effect of ZDV in vivo.

To better assess the antiviral effect of zidovudine (ZDV) in vivo and understand its limitations, we have studied human immunodeficiency virus (HIV) replication in peripheral blood mononuclear cell (PBMC) cultures from 25 ZDV-treated patients and 20 untreated controls. Three months after initiation of therapy, 9 of the 25 treated cases became negative for HIV isolation (36%). Untreated cases never converted to a negative culture. In patients treated by ZDV and who remained culture positive, the kinetics of HIV replication in PBMC culture was found to vary with time. A statistically significant delay in the production of HIV in PBMC cultures from treated cases could be demonstrated after 3 months of ZDV therapy, when compared with untreated patients. By contrast, in such untreated patients the time required to reach the peak of reverse transcriptase activity in culture decreased during the follow-up period. These changes of in vitro HIV replication kinetics as well as the change to negative culture during ZDV therapy probably reflected the reduction of the number of infected cells in vivo. These results as well as the decrease of p24 antigenemia do indicate that ZDV indeed inhibits HIV replication in vivo. However, the effect of ZDV on HIV replication kinetics in PBMC fails to reach significance at 6 months, suggesting that the antiviral effect of ZDV may decrease over time. Our results suggest that ZDV is most effective when the intensity of HIV replication in vivo is still low.