Real-World Analysis of Weight Gain and Body Mass Index Increase Among Patients with HIV-1 Using Antiretroviral Regimen Containing Tenofovir Alafenamide, Tenofovir Disoproxil Fumarate, or Neither in the United States

Abstract

Background: While some studies among patients with HIV-1 suggest that antiretroviral therapy (ART) regimens containing tenofovir alafenamide (TAF) may be associated with greater weight gain than those not containing TAF, no studies have assessed the relationship between TAF doses and weight change. Objectives: To evaluate weight-related outcomes among patients with HIV-1 in the United States initiating ART containing different nucleoside reverse transcriptase inhibitors and doses. Methods: A retrospective longitudinal study was conducted using Decision Resources Group's electronic medical records (July 17, 2017-March 1, 2020). Adult patients with HIV-1 initiating ART (index date) containing TAF 25 mg, TAF 10 mg, tenofovir disoproxil fumarate (TDF), or neither TAF nor TDF on or after July 17, 2018, were included. Changes in weight and body mass index (BMI) from pre-index to 3, 6, 9, and 12 months post-index were compared between cohorts using mean differences obtained from ordinary least squares models adjusted for baseline characteristics. Time-to-weight and BMI increase ≥5% were compared using Cox models adjusted for baseline characteristics. Results: Among 1652 eligible patients (TAF 25 mg, n=710; TAF 10 mg, n=303; TDF, n=219; non-TAF/TDF, n=420), the majority (83.2%-99.5%) initiated an integrase strand transfer inhibitor, except for the TDF cohort (45.2%). Patients initiating TAF 25 mg had greater weight or BMI increase across all time points compared with patients initiating TAF 10 mg, TDF, or non-TAF/TDF regimens (mean differences in weight or BMI changes between cohorts at 12 months post-index ranged from 0.78 kg [1.72 lb] to 1.34 kg [2.95 lb] and from 0.77 kg/m2 to 1.95 kg/m2, respectively), although findings were not statistically significant for all comparisons. Compared with TAF 25 mg, time-to-weight and BMI increase ≥5% in the other treatment cohorts were longer (hazard ratios ranged from 0.77 to 0.94), although findings were generally not statistically significant. Conclusions: Among a population of patients predominantly initiating integrase strand transfer inhibitors, increases in weight and BMI post-ART initiation were common and appeared to be higher and occur more rapidly among patients receiving TAF 25 mg compared with lower TAF doses or other nucleosides. When considering long-term health consequences, weight gain is an important factor to consider when selecting an ART regimen.

Keywords: HIV; body mass index; electronic health records; observational study; tenofovir alafenamide; tenofovir disoproxil fumarate; weight gain.

Figure 1.. Identification of the Study Population

Abbreviations: ART, antiretroviral therapy; BMI, body mass index; CKD, chronic kidney disease; EMR, electronic medical records; ESRD, end-stage renal disease; HIV, human immunodeficiency virus; ICD-9 CM/ICD-10 CM, International Classification of Disease, Ninth/Tenth Revision, Clinical Modification; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumarate. ^a The index period spanned from July 17, 2018 (date of darunavir/cobicistat/emtricitabine/TAF approval), to October 15, 2019, to allow sufficient follow-up time to observe weight or BMI measurements during the observation period. ^b Continuous clinical activity was defined as the period from the first to last record in the EMR database. ^c For multi-tablet regimens (MTRs) identified as part of the TAF 10 mg, TAF 25 mg, or TDF cohorts, the index date was the date of the prescription for the TAF 10 mg, TAF 25 mg, or TDF agent used as part of the MTR. For MTRs identified as part of the non-TAF/TDF cohort, the index date was the date that regimen identification was complete.

Figure 2.. Comparison of Mean Weight or BMI Change Between Pre-index and Post-index Periods

* P value significant at 5% level. Abbreviations: BMI, body mass index; CI, confidence interval; MD, mean difference; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumarate. ^a Mean differences and their associated 95% CIs and P values were estimated using ordinary least squares regression models adjusted for the following baseline characteristics: age, gender, race, insurance plan type, US geographic region, year of the index date, number of mental health–related comorbidities, symptomatic HIV/AIDS, Quan-Charlson Comorbidity Index score, hypertension, type 2 diabetes, prediabetes, baseline BMI, use of an INSTI agent in the index regimen, and use of a medication associated with weight change. A mean difference <0 indicates that the TAF 10 mg, TDF, or non-TAF/TDF cohorts had a lower mean weight gain or BMI increase than the TAF 25 mg cohort.

Figure 3.. Comparison of Patients Having Any, ≥5%, Or ≥10% Weight or BMI Increases Between Pre-index and Post-index Periods

* P value significant at the 5% level. Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumarate. ^a ORs and their associated 95% CIs and P values were estimated using logistic regression models adjusted for the following baseline characteristics: age, gender, race, insurance plan type, US geographic region, year of the index date, number of mental health–related comorbidities, symptomatic HIV/AIDS, Quan-Charlson Comorbidity Index score, hypertension, type 2 diabetes, prediabetes, baseline BMI, use of an INSTI agent in the index regimen, and use of a medication associated with weight change. An OR <1 indicates that the TAF 10 mg, TDF, or non-TAF/TDF cohorts had a lower risk of a weight gain or BMI increase than the TAF 25 mg cohort.

Figure 4.. Kaplan-Meier Curves of Time to Weight Gain or BMI Increase ≥5% or ≥10%

Abbreviations: BMI, body mass index; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumarate.

Figure 5.. Comparison of Time to Weight Gain or BMI Increase ≥5% or ≥10%

* P value significant at the 5% level. Abbreviations: BMI, body mass index; CI, confidence interval; HR, hazards ratio; TAF, tenofovir alafenamide; TDF, tenofovir disoproxil fumarate. ^a HRs and their associated 95% CIs and P values were estimated using Cox proportional hazards regression adjusted for the following baseline characteristics: age, gender, race, insurance plan type, US geographic region, year of the index date, number of mental health-related comorbidities, symptomatic HIV/AIDS, Quan-Charlson Comorbidity Index score, hypertension, type 2 diabetes, prediabetes, baseline BMI, use of an INSTI agent in the index regimen, and use of a medication associated with weight change. An HR <1 indicates that the TAF 10 mg, TDF, or non-TAF/TDF cohorts had a lower risk of weight gain or BMI increase than the TAF 25 mg cohort.