The number of single tablet regimens (STRs) available to clinicians has increased notably since 2011. Few studies have assessed the characteristics of these patients at the time of ART initiation, by formulation. We used EMR data to assess the 4 most widely used STRs.
HIV+ ART-naïve patients initiating a STR between 1/1/2007 and 3/31/2015 were identified in the OPERA database, a collaboration of HIV caregivers at 79 clinics in 15 states. Patients were followed from treatment start to regimen change, death, loss to follow-up, or study end. Demographic and clinical characteristics at time of ART initiation were compared.
A total of 5544 patients initiated an STR (47% efavirenz (EFV)/emtricitabine (FTC)/tenofovir (TDF), 21% rilpivirine (RPV)/FTC/TDF, 29% elvitegravir (EVG)/FTC/TDF/cobicistat (c), 4% dolutegravir (DTG)/abacavir (ABC)/ lamivudine (3TC)). There were significant differences in demographics, route of infection, length of follow-up and baseline VL, CD4, non-HIV pill burden, and comorbidity. Patients initiating DTG/ABC/3TC were more likely to be Hispanic, to initiate with CD4 count ≤50 cells/μL, and less likely to be aged 26-49 or MSM. Patients initiating RPV/FTC/TDF were more likely to be female, African American, to present early and least likely to initiate with VL≥100,000 copies/mL. Patients initiating EFV/FTC/TDF were more likely to be ages 26-49, MSM, HCV+, to present with AIDS and least likely to be African American. Patients initiating EVG/FTD/TDF/c were most likely to reside in the South and Northeast. Median (IQR) follow-up ranged from 13.8 (12.4, 15.5) months on DTG/ABC/3TC to 20.7 (9.3, 41.8) on EFV/FTC/TDF. Based on baseline clinical characteristics prior to start of therapy, median (IQR) VACS index, a measure that predicts all-cause mortality, was highest in DTG/ABC/3TC 23(13, 35) and lowest in RPV/FTC/TDF 17 (13, 23) patients. Differences were sensitive to study time frame since 70% of EFV/FTC/TDF initiated prior to 2011.
Numerous differences in baseline characteristics were observed in STR initiators prior to initiation. If prognostic factors are channeling patients into specific therapies, selection bias may confound outcome evaluation and require the use of advanced methods.
K. Schulman, ViiV Healthcare: Consultant and Research Contractor , Consulting fee and Research grant
C. Henegar, ViiV Healthcare: Research Contractor , Research support
S. Zelt, Viiv Healthcare: Employee , Salary and Stock options
R. D' Amico, ViiV Healthcare: Employee , Salary and stock options
P. Lackey, None