More recent randomized trials found that nevirapine and efavirenz showed similar efficacy [23,24]. In addition, the Atazanavir/Ritonavir

on a background of Tenofovir and Emtricitabine (Truvada) versus Nevirapine (ARTEN) study [25] demonstrated noninferiority between nevirapine and atazanavir, a ritonavir-boosted PI, in a population of antiretroviral-naïve patients. The definition of treatment failure in the 2NN clinical trial [23] was a combined endpoint of virological failure, disease progression or therapy change and the main reason given for treatment failure was a change in therapy. Annan et al. [24] defined treatment failure as either virological failure or discontinuation of therapy. Our analyses were based on reported reason for discontinuation of treatment, rather than treatment failure defined 5-FU in vivo using virological or immunological

measurements, in patients who had initially tolerated and responded to treatment. This definition is closer to the definition of treatment failure used in the more recent studies and our results are consistent with their findings. It has previously been reported that the choice of NRTI backbone is a significant predictor of virological success and treatment failure [24]; however, even after adjustment for this, significant differences remained. In patients with extensive resistance to other drug classes, nevirapine has been associated with an inferior virological outcome

compared ALK tumor with patients on efavirenz [26], and therefore accumulation of resistance from previous drug regimens could also affect the rate of discontinuation because of treatment failure. Around 36% of the patients included in the analysis were treatment naïve at the time of starting their regimen. In naïve patients very few discontinuations, in any group, were because of reported treatment failure. Therefore, in treatment-naïve patients, our results suggest that, if the regimen can be successfully tolerated in the first few months and viral suppression achieved, nevirapine is a durable treatment strategy, in terms of discontinuation because of treatment failure, compared with efavirenz and lopinavir. Patients on lopinavir and efavirenz had a higher rate of discontinuation because of toxicities or patient/physician choice. Other studies found that nevirapine Loperamide was associated with a higher rate of toxicities when compared with efavirenz [1,23] and the ARTEN study [25] found that discontinuation was higher in those on nevirapine compared with atazanavir. However, most of the discontinuations because of toxicity in nevirapine have been reported in the first few months on therapy [16,20,25]. As mentioned previously this analysis focused on patients who had tolerated the first 3 months of therapy. Thus, short-term toxicities, such as hypersensitivity, leading to early discontinuation would have been excluded. Lodwick et al.