Co-receptor switch during HAART is independent of virological success.

The influence of antiretroviral therapy on co-receptor tropism remains controversial. To verify if co-receptor tropism shift was affected by HAART, the evolution of proviral DNA V3 genotype after 12 months of a new antiretroviral regimen was compared between responder and non-responder patients. Baseline blood samples were collected from 36 patients infected with HIV-1 subtype-B (18 naïve and 18 experienced) for virus isolation and env V3 genotyping from plasma HIV-1 RNA and PBMC DNA. DNA V3 genotyping was repeated after 12 months from initiating HAART. WebPSSM was used for categorizing V3 sequences into X4 or R5; for analysis purposes, dual/mixed viruses were considered as X4. From the 10 (28%) patients changing their proviral DNA V3 genotype during therapy, six shifted from R5-to-X4 and four from X4-to-R5. The lack of reaching virological suppression was not associated with an X4-to-R5 (P = 0.25) or R5-to-X4 (P = 0.14) shift; time-to-viral suppression and CD4 increase were similar in both groups. No association was found between tropism shift and patient baseline characteristics including age, sex, CDC stage, CD4 count, viral load, exposure and length of previous HAART, enfuvirtide use in the new regimen, number of reverse transcriptase and protease resistance-associated mutations. Conversely, CD4 nadir was correlated to emergence of X4 virus in proviral DNA (mean 27.2 +/- 30.6 in R5-to-X4 shifting patients vs. 161.6 +/- 150.6 in non-shifting patients, P = 0.02). The occurrence of a tropism shift in both directions was independent of HAART use, irrespective of its efficacy. The CD4 count nadir was the only baseline characteristic able to predict an R5-to-X4 viral shift.