Etravirine in Treatment-Experienced, HIV-1-Infected Children and Adolescents

48-Week Safety, Efficacy and Resistance Analysis of the Phase II PIANO Study

G Tudor-Williams; P Cahn; K Chokephaibulkit; J Fourie; C Karatzios; S Dincq; M Opsomer; TN Kakuda; S Nijs; L Tambuyzer; FL Tomaka


HIV Medicine. 2014;15(9):513-524. 

In This Article


Although therapy for paediatric HIV infection has steadily improved with the introduction of potent antiretroviral combinations, drug choices remain more limited in children than in adults,[1,3,4] and there remains a need for additional antiretroviral options, including age-appropriate formulations. Based on the presented data, etravirine provides an additional treatment option for treatment-experienced, paediatric patients. Etravirine administered at a weight-based dose in combination with an OBR demonstrated safety and efficacy in this difficult-to-treat paediatric population who had VF on their current antiretroviral regimen.

Safety results from this week 48 analysis were comparable to those of week 24 analyses,[23] with no new safety concerns emerging compared with the known safety profile of etravirine in HIV-1-infected adults.[19,20,22,23] Most AEs were grade ≤ 2 and the incidence of serious AEs, grade ≥ 3 AEs or AEs leading to treatment discontinuation was generally low. In addition, mean changes over time in clinical laboratory parameters were small in both age groups and were generally not considered to be clinically relevant.

The most common treatment-related AE was rash (grouped term), reported in 18% of the overall study population. Cases of rash were usually grade ≤ 2 in severity, mostly self limited and usually did not lead to discontinuation: 4.0% of patients discontinued because of rash AEs. The incidence of any rash was similar to that occurring in the etravirine group in adults (19%).[19,20,22,23] In the current study, the median time to first onset was 10 days, and the median duration for rash cases during treatment was 7 days, consistent with data in adults where rash occurred primarily in the second week of treatment. As in adults, rash occurred more frequently in female than in male patients.[19,22] A higher proportion (28%) of participants over 12 years of age were documented as having any rash compared with 15% of children between 6 and 12 years. The study was not powered to enable statistically valid comparisons between arms and this difference may have been a consequence of small numbers. A multivariate analysis on pooled PIANO and DUET data showed that female sex but not age or participation in the PIANO study was a risk factor for developing rash.

Virological response increased progressively over time in both age groups, with sensitivity analyses showing that these results were robust. The overall virological response in this study [56.4% (NC=F)] is in agreement with previous findings with etravirine in HIV-1-infected, treatment-experienced adults.[19,20,22,23] In the phase III DUET trials in adults reusing or not using enfuvirtide, etravirine combined with an OBR resulted in 57% of patients achieving a VL < 50 copies/mL at week 48. In addition, the CD4 count increase (156 cells/μL increase compared with baseline) was consistent with findings in adults.[19,20,22,23] The efficacy of etravirine in this study was comparable with that for NNRTIs currently licensed for paediatric use. Previous trials investigating the use of efavirenz[35,36] and nevirapine[13–15,36] in HIV-infected, treatment-experienced children demonstrated sustained virological responses of > 60%. However, in these trials, patients were NNRTI-naïve.

A multivariate logistic regression model demonstrated potential prognostic factors of virological response to be level of adherence, sex, etravirine weighted genotypic score at baseline and etravirine C0h. However, it should be recognized that the numbers of patients within some subgroups in the multivariate model were small, and results from the subgroup analysis should therefore be interpreted with caution. Multivariate analyses in treatment-experienced adult patients receiving etravirine also identified adherence, but not sex, as predictive of response.[19,20,22,23] In our study, the sex effect on virological response was influenced by the fact that patients who discontinued because of AEs were counted as nonresponders. There were four patients who discontinued for rash, all female. Hence, the effect that was seen was not solely attributable to efficacy, but also attributable to safety.

Although the study was not designed to make comparisons between the two age groups, it should be noted that the virological response at week 48 was lower in adolescents (48.3%) than in children (68.3%). We considered that this was probably attributable to lower adherence, more advanced HIV disease at baseline (higher VL and lower CD4 cell count) and longer treatment history in adolescents compared with children. Previous findings have shown adherence to antiretroviral therapy to be lower in adolescent populations than in children,[37] with pill burden, regimen fatigue, lifestyle issues and lack of knowledge about antiretroviral treatments being barriers to maintaining successful virological control in adolescents.[38,39] Guidelines therefore recommend that strategies to maximize adherence should be discussed before starting therapy or changing regimens.[4]

Measuring adherence in clinical trails is a challenge in the absence of an internationally agreed 'gold standard'. Therapeutic drug monitoring may provide the most direct evidence but even this is subject to misinterpretation if, for example, double drug doses are taken before clinic attendances by otherwise poorly adherent participants. Pill counts and a questionnaire method were used in the PIANO study with discrepant estimates (39% by pill counts; 61–69% by PENTA questionnaire). Given the acknowledged errors associated with each method and the limited size of the study, some discrepancy is to be expected. The questionnaire results are consistent with the VL responses. Actual adherence rates may lie somewhere in between these estimates. Our data confirm the experience from both clinical trials and clinical practice that adherence can be a problem in this age group, and particularly in a selected population who had experienced VF.

Pharmacokinetic data from the current study indicate that etravirine exposure in treatment-experienced children and adolescents was comparable to that seen in adults in the DUET studies [mean (standard deviation) etravirine AUC0–12h and C0h were 5506 (4710) ng h/mL and 393 (391) ng/mL, respectively].[29] These week 48 data therefore substantiate the selection of etravirine 5.2 mg/kg bid dosing for use in paediatric patients.[18]

Although our single-arm study is limited by its small size, given the consistency of our findings with those in adults, we believe that our conclusions are valid. Etravirine is currently being studied in a continued access programme in patients aged 6 to < 18 years (TMC125-C239; NCT00980538). Other studies examining future developments for etravirine paediatric use include the International Maternal Pediatric Adolescent AIDS Clinical Trials (IMPAACT) study (P1090/TMC125-C234; NCT01504841), which will investigate etravirine in children ≥ 2 months to < 6 years old.

In conclusion, the final week 48 analysis data from PIANO have demonstrated that the safety, efficacy and tolerability of etravirine demonstrated at week 24 were sustained to 48 weeks. Etravirine, with activity in patients with resistance to the first-generation NNRTIs, and a favourable lipid and neuropsychiatric safety profile,[40–43] is a useful addition to the NNRTI class as a treatment option for NNRTI treatment-experienced, HIV-1-infected children and adolescents aged 6 to < 18 years.