The Rate of Bone Loss Slows After 1–2 Years of Initial Antiretroviral Therapy

Final Results of the Strategic Timing of Antiretroviral Therapy (START) Bone Mineral Density Substudy

A Carr; B Grund; AV Schwartz; A Avihingsanon; S Badal-Faesen; JI Bernadino; V Estrada; A La Rosa; PWG Mallon; S Pujari; D White; N Wyman Engen; K Ensrud; JF Hoy


HIV Medicine. 2020;21(1):64-70. 

In This Article


Study Design and Participants

Details regarding the START trial's design, participants and assessments have been published previously,[12] as have the baseline characteristics, assessments, endpoints, sample size calculations and statistical plan of its BMD substudy.[13]

START randomized 4684 HIV-positive, ART-naïve adults to immediate ART initiation versus deferring ART. ART regimens were not protocol-specified, but were selected pre-randomization. All remaining untreated participants in the deferred group were offered ART in May 2015 after mean follow-up of 3.0 years as a consequence of demonstration of because significant clinical benefit was demonstrated for immediate ART.[12]

The START BMD substudy co-enrolled 424 START participants at 33 clinical sites in 11 countries between June 2011 and June 2013, with follow-up completed on 31 December 2016. Participants underwent annual dual-energy X-ray absorptiometry (DXA) on a single Hologic (Marlborough, MA, USA) or Lunar scanner (GE Healthcare, Chicago, IL, USA) to measure BMD at the hip and lumbar spine (L1–L4); all scans were performed using a standardized protocol and were centrally analysed.[11]

The substudy was approved by the institutional review board at each participating site and performed in compliance with the Declaration of Helsinki and local regulatory requirements. All participants provided written, informed consent prior to enrolment.

Study Outcomes

The co-primary outcomes were the per cent changes from baseline in total hip BMD and lumbar spine BMD. Pre-specified secondary outcomes included change in femoral neck BMD, and rates of BMD loss upon ART initiation in the immediate ART group and among participants in the deferred group prior to ART start (patients with untreated HIV infection). Evaluation of clinical parameters associated with rates of BMD change was also planned. Fractures were reported for all participants in the parent START study at baseline and annually.

Statistical Analyses

Changes in BMD from baseline were expressed as per cent of baseline BMD. The primary analysis was the intent-to-treat (ITT) comparison between the immediate and deferred ART groups for per cent change in BMD using longitudinal mixed models, adjusted for baseline BMD and visit. Groups were compared for changes in BMD to each year of follow-up using analysis of covariance (ANCOVA) models adjusted for baseline BMD. We also compared the immediate group (excluding those who did not start ART during year 1) versus the deferred group censored at ART start. Methods to estimate annual rates of BMD change, for subgroup analyses, and to estimate associations of baseline factors with changes in BMD were described previously.[11]

Fracture incidence rates were estimated in the parent START population using data acquired up to 31 December 2016. Treatment groups were compared by ITT using a Cox proportional hazards model, stratified by region.

Analyses were performed with SAS version 9.3 (SAS Institute, Cary, NC) and R version 3.[14]