Home Search Abstracts View Session E-mail Abstract Author


Session 13-Themed Discussion
TD: NeuroAIDS Treatment Issues and Controversies
Wednesday, 1-2 pm; Room 2010
Paper # 432    
CSF Escape Is Uncommon in HIV-1-infected Patients on Stable ART
Arvid Edén1, R Price2, L Hagberg1, and M Gisslén1
1The Sahlgrenska Academy at Univ of Gothenburg, Sahlgrenska Univ Hosp, Sweden and 2Univ of California, San Francisco and San Francisco Gen Hosp, US

Background:  ART effectively reduces HIV-1 RNA in plasma, as well as in cerebrospinal fluid (CSF). Occasionally, patients exhibit CSF escape, or elevated RNA levels in CSF while effectively suppressed in plasma. Here, we examine CSF escape in patients effectively treated with ART in relation to treatment regimens.

Methods:  There were 63 patients from 2 centers treated with ART regimens of either efavirenz, lopinavir/ritonavir or atazanavir/ritonavir in combination with 2 NRTI’s consisting of lamivudine or emtricitabine with either tenofovir, abacavir, or zidovudine for at least 6 months, with undetectable (<50 copies/mL) HIV-1 RNA in plasma were included in the analysis. HIV-1 RNA was analysed with Roche Amplicore v.1.5 or Cobas Taqman v.1.0. Fisher’s exact test was used for statistical analysis.

Results:  Of 63 identified patients, 7 (11 %) had signs of CSF escape with detectable HIV-1 RNA levels in CSF (range: 46 to 213 copies/mL) despite having undetectable levels in blood. Five of 24 (21 %) efavirenz-treated and 2/13 (15 %) atazanavir-treated patients had detectable RNA in CSF. Of 26 lopinavir-treated patients, none had detectable CSF-RNA, with significantly lower frequency of CSF escape compared to efavirenz-treated patients (=0.02). Other differences between treatment regimens did not reach statistical significance; 3/17 (18 %) and 4/27 (15 %) of patients treated with abacavir and tenofovir had detectable CSF RNA, respectively. Nineteen patients received a NRTI backbone containing zidovudine, none of which had detectable CSF RNA.

Conclusions:  CSF escape, measured as detectable CSF HIV-1 RNA, remains an uncommon finding in patients with successful suppression of plasma viral load. However, there may be cause for consideration of specific ART regimen regarding CNS efficacy, especially in patients with HIV-1 related neurological disease. In our material, CSF escape was most common in patients treated with efavirenz in combination with either abacavir or tenofovir. Of note, no patients treated with the PI lopinavir/ritonavir, or the NRTI zidovudine, had detectable CSF HIV-1 RNA, while retaining undetectable plasma viral load.