Metabolic Effects of Preexposure Prophylaxis With Coformulated Tenofovir Disoproxil Fumarate and Emtricitabine

Abstract

Background: Antiretroviral drugs have been associated with changes in lipids, fat mass and dat distribution. Tenofovir disoproxil fumarate (TDF) has been shown to have a more favorable metabolic profile than other drugs in its class. However, the metabolic effects of TDF in preexposure prophylaxis (PrEP) are unknown.

Methods: We evaluated the effects of TDF/emtricitabine (FTC) on lipids and body composition in a blinded, placebo-controlled PrEP trial. Participants enrolled in a metabolic subcohort (N = 251, TDF/FTC; N = 247, placebo) consented to fasting lipid panels, dual-energy X-ray absorptiometry scans for body composition, and pharmacologic testing of drug metabolites at baseline and every 24 weeks thereafter.

Results: Lean body mass was stable and unaffected by TDF/FTC. Body weight increased in both groups but was lower on TDF/FTC through week 72. This difference was explained by lower fat accumulation on TDF/FTC. The net median percent difference (standard error, P value) for TDF/FTC vs placebo at week 24 was -0.8% (0.4%, P = .02), +0.3% (0.4%, P = .46), and -3.8% (1.4%, P = .009) for total, lean, and fat mass, respectively. There was no apparent differential regional fat accumulation on TDF/FTC. Decreases in cholesterol, but not triglycerides, were seen in TDF/FTC participants, with detectable drug levels compared to placebo.

Conclusions: TDF/FTC for PrEP showed cholesterol reductions and appeared to transiently suppress the accumulation of weight and body fat compared to placebo. There was no evidence of altered fat distribution or lipodystrophy during daily oral TDF/FTC PrEP.

Clinical trials registration: NCT00458393.

Figure 1.

Median percentage increase in body weight in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with at least 1 dual-energy X-ray absorptiometry scan available in each visit window. Multiple scans by individuals in windows were averaged. x-axis (panels A and B): weeks since randomization. y-axis (panels A and B): median percent change in body weight from baseline. Abbreviations: BLQ, below level of quantitation; FTC, emtricitabine; TDF, tenofovir disoproxil fumarate.

Figure 2.

A, Median percentage increase in lean body mass as measured by dual X-ray absorptiometry (DXA) in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with at least 1 DXA scan available in each visit window. Multiple scans by individuals in windows were averaged. B, Median percentage increase in total fat mass as measured by dual X-ray absorptiometry (DXA) in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with at least 1 DXA scan available in each visit window. Multiple scans by individuals in windows were averaged. x-axis (panels A and B): weeks since randomization. y-axis (panel A): median percent change in lean body mass from baseline. y-axis (panel B): median percent change in fat mass from baseline Abbreviations: BLQ, below level of quantitation; FTC, emtricitabine; TDF, tenofovir disoproxil fumarate.

Figure 3.

A, Median percentage increase in fat mass in the trunk as measured by dual X-ray absorptiometry (DXA) in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with at least 1 DXA scan available in each visit window. Multiple scans by individuals in windows were averaged. B, Median percentage increase in fat mass in the limbs (arms and legs) as measured by dual X-ray absorptiometry (DXA) in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with at least 1 DXA scan available in each visit window. Multiple scans by individuals in windows were averaged. x-axis (panels A and B): weeks since randomization. y-axis (panel A): median percent change in trunk fat mass from baseline. y-axis (panel B): median percent change in limb fat mass from baseline Abbreviations: BLQ, below the level of quantitation; FTC, emtricitabine; TDF, tenofovir disoproxil fumarate.

Figure 4.

A, Median percentage increase in fasting total cholesterol in the metabolic cohort+. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with \ lipid results available in each visit window. Multiple results by individuals in windows were averaged. B, Median percentage increase in fasting low-density lipoprotein (LDL) in the metabolic cohort+. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with lipid results available in each visit window. Multiple results by individuals in windows were averaged. C, Median percentage increase in fasting high-density lipoprotein (HDL) in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with lipid results available in each visit window. Multiple results by individuals in windows were averaged. D, Median percentage increase in triglycerides in the metabolic cohort. Curves are stratified by randomized treatment (placebo vs tenofovir disoproxil fumarate/emtricitabine [TDF/FTC]) with the TDF/FTC group further stratified by detection of either tenofovir or FTC in a plasma sample (“Detected”) from that visit or not (“BLQ”). Drug level testing was conducted through week 72. Bars indicate 1 standard error. The table below the axis gives the number of participants per group with lipid results available in each visit window. Multiple results by individuals in windows were averaged. x-axis (panels A–D): weeks since randomization. y-axis (panel A): median percent change in total cholesterol from baseline. y-axis (panel B): median percent change in LDL-C from baseline. y-axis (panel A): median percent change in HDL-C from baseline. y-axis (panel D): median percent change in triglycerides from baseline. Abbreviations: BLQ, below the level of quantitation; FTC, emtricitabine; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TDF, tenofovir disoproxil fumarate.