Incidence of Dyslipidemia in People With HIV Who Are Treated With Integrase Inhibitors Versus Other Antiretroviral Agents: The RESPOND Study Group

Dathan M. Byonanebye; Mark N. Polizzotto; Josip Begovac; Katharina Grabmeier-Pfistershammer; Irene Abela; Antonella Castagna; Stéphane DeWit; Cristina Mussini; Jörg J. Vehreschild; Antonella d'A Monforte; Ferdinand W.N.M. Wit; Christian Pradier; Nikoloz Chkhartishvili; Anders Sönnerborg; Jennifer Hoy; Jens Lundgren; Bastian Neesgaard; Loveleen Bansi-matharu; Lauren Greenberg; Josep M. Llibre; Vani Vannappagari; Joel Gallant; Coca Necsoi; Piotr Cichon; Peter Reiss; Inka Aho; Tengiz Tsertsvadze; Marianna Mennozzi; Andri Rauch; Camilla Muccini; Matthew Law; Amanda Mocroft; Lene Ryom; Kathy Petoumenos

Disclosures

AIDS. 2021;35(6):869-882. 

In This Article

Results

Baseline Characteristics

At baseline, 7886 participants had lipid measures, 3309 (42.0%) of whom had dyslipidemia at baseline and were excluded. The participants with dyslipidemia at baseline were older and more likely to have hypertension, diabetes mellitus, CKD, and CVD than those with no dyslipidemia (Supplementary Table S1, http://links.lww.com/QAD/B984). Overall, 4577 (58.0%) PLWH were eligible for the primary analysis, of whom 3063 (66.9%) were on INSTI (DTG = 1757, RAL = 422, EVG/c = 884), 571 (12.5%) on PI/b (ATV/b = 187, DRV/b = 384), and 943 (20.6%) on NNRTI (RPV = 618, EFV = 325). Most participants were male (72.4%), had prior ART exposure (57.7%), and were of white ethnicity (72.5%) (Table 1). The median age [interquartile range (IQR)] was 43 (35–51) years.

Compared with individuals treated with NNRTI or PI/b, participants taking INSTI were older, more ART-experienced and more likely to be hepatitis B and C positive, have hypertension, CVD, higher baseline TC, LDL, and TRIGs levels, baseline CD4+ cell count, lower nadir CD4+ cell count, and more likely to be smokers. Conversely, PLWH on PI/b had a higher lipid-testing frequency compared with those taking INSTI and NNRTI. These differences were broadly consistent in the ART-naive and ART-experienced PLWH subgroups. The most common NRTI backbones in the initiated ART regimens were TDF/emtricitabine (n = 2449; 53.5%), abacavir/lamivudine (n = 1330; 29.1%) and TAF/emtricitabine (n = 683;14.9%). In the participants who switched to INSTI, the switched regimen was mainly protease inhibitors (56.2%) or NNRTI (34.8%). The switched regimen was protease inhibitors in 83.6% of participants who switched to NNRTI and 78.0% NNRTI in those who switched to protease inhibitors.

Incidence of Dyslipidemia

In the primary analysis, 1460 of 4577 (32.0%) participants developed dyslipidemia during 7621.3 person-years of follow-up (PYFU) [incidence rate, 191.6 per 1000 PYFU; 95% confidence intervals (CIs), 182.0–201.7]. Of these, 407 (27.9%) had more than one abnormal lipid measure or initiated lipid-lowering therapy, 207 (14.2%) had only hypercholesteremia, 407 (27.9%) had only hypertriglyceridemia, 396 (27.1%) had low HDL, and 43 (2.9%) initiated statin without abnormal lipid measures. The median follow-up period (IQR) was 1.7 (0.6–3.0) years. Participants taking INSTI had a 29% lower incidence of dyslipidemia compared with those taking PI/b (aIRR 0.71; CI 0.59–0.85; P < 0.001). Conversely, INSTI-treated participants had a 35% higher rate of dyslipidemia compared with individuals taking NNRTI (aIRR 1.35; CI 1.16–1.58; P < 0.001) (Figure 1).

Figure 1.

Overall adjusted and nonadjusted incidence rate ratios of dyslipidemia in people with HIV taking different antiretroviral therapy regimens versus integrase inhibitors. aAll PLWH means both ART naïve and ART-experienced individuals; Integrase inhibitors; NNRTI-Non-nucleoside reverse transcriptase inhibitors; PI boosted protease inhibitors; bAdjustment was done for baseline lipid values as well as age, ethnicity, gender, mode of HIV transmission, baseline calendar year, smoking status, BMI, hypertension, diabetes mellitus, prior AIDS, CVD, CKD, HBV, HCV, HIV RNA, nadir and baseline CD4 counts, duration of HIV positive status, and prior ART exposure (in the analysis for all PLWH). A risk ratio (RR) > 1 means that the ART regimen was associated with a higher incidence of dyslipidemia compared to INSTI; while a risk ratio < 1 means that the regimen was associated with a lower incidence of dyslipidemia compared to INSTI.

In the final adjusted model, other covariates that were associated with a lower incidence of dyslipidemia were female sex (aIRR 0.80; CI 0.70–0.92; P = 0.002), black ethnicity (aIRR 0.69; CI 0.56–0.86; P = 0.001), prior ART exposure (aIRR 0.74; CI 0.65–0.85; P < 0.001), and commencement of ART after the year 2013 (Table 2). Conversely, the predictors of dyslipidemia were baseline CD4+ cell count less than 350 cells/μl (aIRR 1.29, compared with CD4+ cell count ≥500 cell/μl; CI 1.13–1.48; P < 0.001), diabetes mellitus (aIRR 1.35; CI 1.05–1.74; P = 0.020), and BMI more than 25 kg/m2 (aIRR 1.14; CI 1.01–1.30; P = 0.045) and higher baseline lipid levels (TC ≥ 200 mg/dl and/or LDL ≥ 100 mg/dl and/or TRIGs ≥ 150 mg/dl), and HDL less than 40 mg/dl. The risk of dyslipidemia increased by 5% (CI 2–7%; P < 0.001) for every 5-year increase in age (Table 2). Prior AIDS, hepatitis C, smoking, hypertension, CKD, HIV RNA at least 200 copies/ml, and longer HIV duration were not associated with a higher incidence of dyslipidemia (Supplementary Table S2, http://links.lww.com/QAD/B984).

Incidence of Dyslipidemia in Antiretroviral Therapy-naive People With HIV

Among the ART-naive participants (n = 1938), 682 (35.2%) developed dyslipidemia during 2995.1 PYFU: incidence rate (per 1000 PYFU) 227.7 (CI 211.2–245.5). The incidence of dyslipidemia was 40% lower in individuals taking INSTI than in those taking PI/b (aIRR 0.60; CI 0.50–0.72; P < 0.001). The rates of dyslipidemia in individuals treated with INSTI were not significantly different from those taking NNRTI (aIRR 1.12; CI 0.93–1.36; P = 0.245) (Figure 1). After controlling for baseline lipid levels, dyslipidemia was more common in participants with BMI greater than normal (>25 kg/m2), prior severe immunosuppression (nadir CD4+ cell count <200 cells/μl), and older age (Supplementary Table S3, http://links.lww.com/QAD/B984).

Incidence of Dyslipidemia in Antiretroviral Therapy-experienced People With HIV

In ART-experienced PLWH (n = 2639), 778 (29.5%) developed dyslipidemia during 4626.1 PYFU: incidence rate (per 1000 PYFU) 168.2 (CI 156.8–180.4). The incidence of dyslipidemia was 43% lower in individuals taking INSTI than in those taking PI/b (aIRR 0.57; CI 0.41–0.81; P = 0.002). The incidence was 36% higher in individuals taking INSTI than in those taking NNRTI (aIRR 1.36; CI 1.08–1.71; P < 0.001) (Supplementary Table S4, http://links.lww.com/QAD/B984).

Incidence of Dyslipidemia Using the ATP III Criteria

Overall, 1560 (36.7%) of the 4252 participants who had no dyslipidemia using the ATP III criteria developed dyslipidemia during 6732.3 PYFU: incidence rate 231.7 per 1000 PYFU (CI 220.5–243.5). The incidence of dyslipidemia in individuals taking INSTI was lower than that in those taking PI/b (aIRR 0.66; CI 0.55–0.79; P < 0.001), but higher than in individuals taking NNRTI (aIRR 1.29; CI 1.11–1.50; P = 0.001). In the final adjusted model (Supplementary Table S5, http://links.lww.com/QAD/B984), diabetes mellitus, BMI at least 25 kg/m2, prior AIDS, and current or prior smoking were associated with a higher incidence of dyslipidemia. The incidence of dyslipidemia increased by 3% (CI 1–6%) for every 5-year increase in age. Conversely, female sex, black ethnicity, and prior ART exposure conferred a lower risk.

Incidence of Overall and Serious Lipid Abnormalities

During follow-up, 1699 participants developed lipid abnormalities (DAIDS Grade ≥1) during 7806.1 PYFU (incidence rate 217.7 per 1000 PYFU; CI 207.5–228.2), and 318 serious lipid abnormalities during 16 018.4 PYFU (incidence rate 19.9 per 1000 PYFU; CI 17.8–22.2). As shown in Figure 2, the rates of lipid abnormalities were lower in INSTI-treated than in protease inhibitors-treated individuals (all lipid abnormalities: aIRR 0.76; CI 0.66–0.87; P < 0.001; serious abnormalities: aIRR 0.57; CI 0.41–0.80; P = 0.001). The incidence of all grades of lipid abnormalities was higher in individuals taking INSTI than in those taking NNRTI (aIRR 1.30; CI 1.14–1.49; P < 0.001). However, the rates of lipid serious abnormalities did not differ between participants taking INSTI or NNRTI (aIRR 1.34; CI 0.95–1.87; P = 0.091).

Figure 2.

Adjusted incidence rate ratios of overall and serious lipid abnormalities in people with HIV treated with antiretroviral therapy classes versus integrase inhibitors. aINSTI, integrase inhibitors; NNRTI, nonnucleoside reverse transcriptase inhibitors; PI/b, boosted protease inhibitors; bthe adverse events were graded as per the division of AIDS table for grading the severity of adult adverse events [31]. A risk ratio more than 1 means that the regimen was associated with a higher incidence of dyslipidemia compared with integrase inhibitors while a risk ratio less than 1 means that the regimen was associated with a lower incidence of dyslipidemia compared with integrase inhibitors.

Incidence of Dyslipidemia by Specific Antiretroviral Drugs

Compared with DTG, the incidence of dyslipidemia was higher in participants taking EVG/c (aIRR 1.20; CI 1.00–1.43; P = 0.046) and RAL (aIRR 1.24; CI 1.02–1.52; P = 0.029). Elvitegravir/cobicistat and raltegravir were also associated with a higher incidence of abnormal TC and TRIGs, but with lower rates of low HDL. However, this was significant for only EGV/c. However, participants receiving RAL had higher rates of lipid-lowering therapy commencement. Similarly, participants taking ATV/b or DRV/b had a higher incidence of dyslipidemia compared with DTG. DRV/b and ATV/b were also associated with higher rates of abnormal TC and TRIGs levels. On the other hand, the incidence of dyslipidemia was 23% lower in participants taking RPV than in those taking DTG (aIRR 0.77; CI 0.63–0.94; P = 0.009). RPV was associated with a lower risk of abnormal TRIGs and TC levels (Figure 3). Conversely, participants taking EFV had higher rates of abnormal TC and commencement of lipid-lowering therapy.

Figure 3.

Adjusted incidence rate ratios of dyslipidemia in people with HIV taking specific antiretroviral drugs versus dolutegravir. ATV/b, boosted atazanavir; DRV/b-boosted darunavir; DTG, dolutegravir; EFV, efavirenz; EVG/c, cobicistat-boosted elvitegravir; RAL, raltegravir; RPV, rilpivirine. A risk ratio less than 1 means that participants taking antiretroviral therapy consisting of the specific antiretroviral had a lower incidence of dyslipidemia compared with those taking dolutegravir-based antiretroviral therapy while a risk ratio more than 1 means that participants taking antiretroviral therapy consisting of the specific antiretroviral had a higher incidence of dyslipidemia compared with those taking dolutegravir-based antiretroviral therapy.

The incidence of dyslipidemia was not different between participants taking TAF or ABC versus TDF (TAF: aIRR 1.15; CI 0.96–1.38, P = 0.121; ABC: aIRR 1.00; CI 0.88–1.14, P = 0.973). However, participants taking TAF or ABC had a higher incidence of hypercholesterolemia (>240 mg/dl) compared with those taking TDF (TAF: aIRR 1.71; CI 1.37–2.12, P < 0.001; ABC: aIRR 1.50; CI 1.25–1.81; P < 0.001), but only TAF was associated with a higher incidence of hypertriglyceridemia (>200 mg/dl) (aIRR 1.32; CI 1.08–1.61; P = 0.006). Participants taking TAF also had lower rates of low HDL levels (<35 mg/dl) than those taking TDF (aIRR 0.58; CI 0.45–0.75; P < 0.009).

Sensitivity Analysis

In all the sensitivity analyses (Supplementary Table S6, http://links.lww.com/QAD/B984), dyslipidemia was consistently less common in participants taking INSTI than in those taking PI/b. Dyslipidemia rates were also generally higher in individuals taking INSTI than in those taking NNRTI. Compared with participants taking INSTI, those taking PI/b were more likely to develop abnormal individual lipids, except low HDL levels, and commence lipid-lowering therapy (aIRR 1.91; CI 1.24–2.96; P = 0.003).

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....