Home Search Abstracts View Session E-mail Abstract Author


Session 124 Poster Abstracts
Morbidity and Mortality: Non-AIDS Events
Session Day and Time: Monday, 1-2:30 pm
Poster Hall


709
Different Rates of Discontinuation because of Toxicities According to CD4 Counts and Prior ART in Patients Starting Nevirapine-based cART: Nevirapine Toxicity Multi-cohort Collaboration
Anouk Kesselring*1, F Wit2, C Sabin3, J Lundgren4, J Gill5, J Gatell6, A Rauch7, J Montaner8, P Reiss2, and A Mocroft3
1HIV Monitoring Fndn, Amsterdam, The Netherlands; 2Academic Med Ctr, Univ of Amsterdam, The Netherlands; 3Royal Free and Univ Coll Med Sch, London, UK; 4Rigshospitalet and University of Copenhagen, Copenhagen HIV Programme, Copenhagen, Denmark; 5Univ of Calgary, Canada; 6Hosp Clin-IDIBAPS, Univ of Barcelona, Spain; 7University Hospital Bern and University of Bern, Bern, Switzerland; and 8British Columbia Center for Excellence in HIV/AIDS, St. Paul`s Hospital, Vancouver, Canada

Background:  Recommendations to avoid nevirapine (NVP) in patients starting combination ART (cART) at higher CD4 counts are based on studies in treatment-naive patients. Further clarification of risk was limited by a lack of power. This collaborative consortium compared risk factors for toxicities in naive and experienced patients starting NVP-based cART (NVPc).

Methods:  Patients from Athena, British Columbia, EuroSIDA, Hospital Clinic Barcelona, Southern Alberta, Swiss HIV, and UK CHIC Cohort studies starting NVPc after January 1, 1998 were included. CD4 at starting NVPc was classified as high (>400/mm3/>250/mm3 in males/females, respectively) or low. Cox models compared the risk of discontinuation due to toxicities or patient/physician choice (TOXPC, n = 10,186); and discontinuation due to hypersensitivity reactions (HSR, for cohorts with detailed data on reasons for discontinuation, n = 6547). Patients were classified according to prior treatment experience and CD4/viral load at start NVPc. All models were stratified by cohort and adjusted for age, gender, calendar year of starting NVPc, and mode of transmission.

Results:  Overall, 61% of patients were Caucasian, 3% Asian, 27% female, 10% injecting drug users (IDU), median age 38 years (IQR 33 to 45). The median times from start NVPc to TOXPC and HSR were 162 days (IQR 31 to 737) and 30 days (IQR 17 to 60), respectively. In the adjusted Cox analysis experienced patients with high CD4 and viral load >400 had a higher risk for TOXPC 1.4 [1.2 to 1.6], p <0.0001) and HSR (1.8 [1.2 to 2.6], p = 0.002), compared to naive patients with low CD4. Experienced patients with high CD4 and viral load ≤400 had no increased risk (TOXPC 1.0 [0.9 to 1.1], p = 0.7, HSR 1.3 [0.9 to 1.7], p = 0.14). Other factors associated with TOXPC were female gender 1.2 (1.1 to 1.3), p = 0.002 (HSR 1.7 [1.2 to 2.3], p = 0.0009), Asian race 1.3 (1.0 to 1.6), p = 0.05 (HSR 2.0 [1.3 to 3.2], p = 0.002), and starting cART after 2002 1.2 (1.1 to 1.3), p = 0.0001 (HSR 1.5 [1.2 to 1.9], p = 0.0002). Other factors associated with TOXPC, but not HSR were IDU 1.6 (1.4 to 1.9), p <0.0001, and older age per 10 years 0.9 (0.9 to 1.0), p <0.0001.

 

 

 

Conclusions:  Our results suggest that it may be relatively safe to initiate NVPc in ART-experienced patients with high CD4 counts provided there is no detectable viremia. Naive patients with high CD4, experienced patients with high CD4 and detectable viral load, females, and patients from Asian origin have an increased risk for TOXPC and HSR.