Incidence and predictors of regimen-modification from first-line antiretroviral therapy in Thailand: a cohort study

BACKGROUND: Antiretroviral therapy markedly reduced mortality in HIV-infected individuals. However, in the previous studies, up to 50% of patients are compelled to modify their regimen in middle and low-income countries where salvage drug is still limited. This cohort study aimed to investigate the incidence and predictors of regimen modification from the first-line antiretroviral regimen in northern Thailand. METHODS: All HIV-infected patients starting antiretroviral therapy (ART) with generic drug (GPOvir®; stavudine, lamivudine and nevirapine) at a governmental hospital in northern Thailand from 2002 to 2007 were recruited. Baseline characteristics and detailed information of regimen modification until the end of 2010 were ascertained from cohort database and medical charts. As a potential genetic predictor of regimen modification, HLA B allele was determined by bead-based array hybridization (WAKFlow® HLA typing kit). We investigated predictors of the regimen modification using Cox’s proportional hazard models. RESULTS: Of 979 patients, 914 were eligible for the analysis. The observed events of regimen modification was 377, corresponding to an incidence 13.8/100 person-year-observation (95% CI:12.5-15.3) over 2,728 person years (PY) follow up. The main reasons for regimen modification were adverse effects (73.5%), especially lipodystrophy (63.2%) followed by rash (17.7%). Sixty three patients (17.1%) changed the regimen due to treatment failure. 2% and 19% of patients had HLA-B*35:05 and B*4001, respectively. HLA-B*35:05 was independently associated with rash-related regimen modification (aHR 7.73, 95% CI:3.16-18.9) while female gender was associated with lipodystrophy (aHR 2.11, 95% CI:1.51-2.95). Female gender (aHR 0.54, 95% CI: 0.30-0.96), elder age (aHR 0.56, 95% CI: 0.32-0.99) and having HLA-B*40:01 (aHR 0.29, 95% CI: 0.10-0.82) were protective for treatment failure related modification. CONCLUSION: HLA-B*35:05 and female gender were strong predictors of regimen modification due to rash and lipodystrophy, respectively. Female gender, elder age, and having HLA-B*40:01 had protective effects on treatment failure-related regimen modification. This study provides further information of regimen modification for future tailored ART in Asia. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12879-014-0565-5) contains supplementary material, which is available to authorized users.