icon- folder.gif   Conference Reports for NATAP  
  AIDS 2010
18th International AIDS Conference (IAC)
July 18-23 2010
Vienna, Austria
Back grey_arrow_rt.gif
The impact of switching to etravirine on efavirenz-related CNS toxicity
  Reported by Jules Levin
Laura Waters1, Martin Fisher2, Alan Winston3, Chris Higgs1, Wendy Hadley2, Lucy Garvey3, Sundhiya Mandalia1, Nicky Perry2, Nicola Mackie3& Mark Nelson1.St Stephen's AIDS Research, Chelsea & Westminster Hospital, London, UK
Switching patients with CNS AE from EFV to ETR resulted in significant improvements in overall grade 2-4 CNS AE, grade 2-4 insomnia, abnormal dreams, nervousness, median number CNS AE and total CNS score.
ETR provides an effective and tolerable switch option in patients with EFV-related CNS AE but the lack of improvement for some AE and patients highlights the importance of other causes of CNS symptoms
Drug-related adverse events are an important cause of poor adherence, treatment failure and a precipitating factor for switching therapy. Timely switch of antiretrovirals, where alternative options are available, is an important strategy for managing drug-related adverse events.
Efavirenz (EFV) is the preferred agent in the British guidelines. Central nervous system adverse events (CNS AE) are common on EFV-based regimens and although most CNS AE are transient a significant minority of individuals experience ongoing CNS AE
Differentiating drug AE from other causes of CNS problems can be difficult.
Etravirine (ETR) is a second generation NNRTI licensed for treatment-experienced HIV-1 infected patients with a favourable toxicity profile in clinical trials. We have previously shown that switching directly from EFV to ETR is feasible from a pharmacokinetic standpoint.
Our aim was to investigate the impact of switching to ETR in patients with ongoing CNS AE on an EFV-containing HAART regimen.
We performed a randomised, placebo-controlled, double-blind phase 4 study at 3 UK study sites: Chelsea & Westminster (London), Royal Sussex County Hospital (Brighton) and St Mary's Hospital (London).
All patients on EFV with an undetectable plasma HIV-RNA (less than 50 copies/ml) for at least 12 weeks with CNS symptoms (please contact the author for a full list of eligibility criteria). Written informed consent was obtained and ethical approval was granted by Riverside Ethics Committee, London.
Recruited patients were randomised to one of 2 arms as illustrated in the flow diagram below. The 1st 12 weeks were double-blind followed by a 12-week open-label phase.

Twelve CNS AE were considered as per the IAS DAIDS grading scale (dizziness, depression, insomnia, anxiety, pain, impaired concentration, headache, somnolence, fatigue, abnormal dreams, nervousnessand hallucinations).
We considered the proportion of patients with G2-4 CNS AE (overall and individual), median number of G2-4 CNS AE and a CNS score (sum total of all grades of CNS AE eg G1 dizziness + G3 abnormal dreams = CNS score 4). Primary end point was change in proportion of subjects with G2-4 CNS AE from baseline to week 12. Secondary end-points included change in CNS score and median number of CNS AE from baseline to week 12 and week 12 to week 24, combined changed in CNS AE after 12 weeks of ETR (ie a combined changes from baseline to week 12 in the IS arm and week 12 to 24 in the DS arm), lipids, CD4 and HIV-RNA.
Statistical analysis were performed in S.A.S v9.1 using standard qualitative and quantitative tests, adjusted for small numbers; all p-values are 2-tailed (contact author for full details).

38 male patients were enrolled, 37 Caucasian, with a median age of 43 years. All had HIV-RNA <50 copies/ml at baseline and median CD4 was 468 (IQR 378-580) cells/mm3.
Baseline NRTI backbone was as follows: TDF/FTC 29%, TDF/FTC as FDC with EFV 29%, ABC/3TC 29%, ABC/TDF 8% and TDF/3TC 5%. Median duration of prior EFV exposure was 21 months.
Four patients discontinued the study: 1 in the IS arm (LFU) and 3 in the DS arm (2 LFU, 1 AE [admission for viral URTI]).
Baseline CNS AE were similar in the 2 arms with 90% and 88.9% in the IS and DS arms, respectively, reporting at least one G2-4 CNS AE. Baseline insomnia was more common in the IS than DS arm (75% vs 38.9%; p=0.024).
During the blinded phase there was a significant reduction in overall G2-4 CNS AE (90% to 60%; p=0.041) in the IS arm (ie in those patients who switched to ETR) and in G2-4 abnormal dreams (50% to 20%; p=0.041) with a trend to reduction in insomnia (75% to 50%; p=0.074). There were no significant changes in any G2-4 CNS AE from baseline to week 12 in the DS arm (ie those patients who remained on EFV).
From week 12 to week 24 (open label phase) there were no further significant changes on any grade 2-4 CNS AE in the IS arm but a significant reduction in G2-4 abnormal dreams in the DS arm (63% to 20%; p=0.023).
From baseline to week 12, median number of G2-4 CNS AE fell from 4 to 1.5 (p=0.08) in the IS arm and were unchanged at 3 in the DS arm. From week 12 to 24 there was a further reduction in median G2-4 CNS AE of 0.5 in the IS arm (p=ns) and from 3 to 1 in the DS arm (p=0.008).
CNS score declined from 14 to 6 between baseline and week 12 in the IS arm (p=0.001) and from 10 to 7.5 in the DS arm (p=ns). Changes from week 12 to 24 in each arm were non-significant.
Combined analyses (change from baseline to week 12 in IS arm and week 12 to 24 in DS arm) are illustrated below.

Combined CNS score declined from 10 to 7 after 12 weeks of ETR (p=0.001). All patients maintained HIV-RNA less than 50 copies/ml at all visits and CD4 change from baseline to week 24 was +68 and +60 cells/mm3 in the IS and DS arms, respectively.
Switching from EFV to ETR resulted in significant reduction in total & LDL cholesterol with no change in HDL (see table below).

Acknowledgments: this work was supported by an Investigator Initiated Grant from Tibotec, a division of Janssen-Cilag.For copies of poster or additional data please contact laura.waters@btinternet.com