Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

Abstract

Objectives

We aimed to assess the clinical characteristics associated with the use of two-drug regimens (2DRs) and the factors associated with specific antiretrovirals in 2DRs in a large single-centre HIV cohort.

Methods

Retrospective analysis of demographics, HIV characteristics and AIDS events, antiretroviral prescription, virological failure and genotypic resistance testing, and laboratory results from all adult people with HIV (PWH) prospectively followed at the Hospital Clinic of Barcelona who were receiving a 3DR or a 2DR in January 2020. We assessed factors associated with the probability of receiving 2DRs relative to three-drug regimens (3DRs) using a logistic regression model, controlling for age, sex and year of HIV diagnosis. The same methodology was applied to identify factors associated with the prescription of integrase inhibitor-based regimens or PI-based regimens among PWH receiving 2DRs.

Results

There were 3432 (88%) PWH receiving 3DRs and 463 (12%) receiving 2DRs. In the final adjusted model, ≥2 previous virological failures, previous resistance mutations, previous AIDS diagnosis, longer time on current regimen, higher total cholesterol or triglycerides and lower baseline haemoglobin were independent factors associated with 2DRs. The majority of 2DRs included an integrase inhibitor or/and a PI. We identified independent factors associated with the inclusion of integrase inhibitors (lower HDL cholesterol) or PIs (prior AIDS, prior genotypic resistance mutations and lower CD4/CD8 ratio) in the 2DR.

Conclusions

In this large single-centre HIV cohort, a worse cardiometabolic status or more archived resistance were key factors associated with inclusion of integrase inhibitors or PIs, respectively, in 2DRs.

Introduction

Major guidelines have long-established three-drug regimens (3DRs) consisting of two NRTIs plus a third drug as the gold standard of ART.1–3 Antiretroviral drugs have improved over time, becoming more effective, simpler and better tolerated. However, for some patients, standard 3DRs may still be challenging because of antiretroviral-related adverse effects, negative impact on comorbidities, risk of interactions, archived resistance or other reasons, therefore justifying the need for an individualized therapeutic approach with fewer than three drugs.4,5 PI6 and dolutegravir7 monotherapy are inferior to standard triple therapy and are discouraged, but the efficacy of two-drug regimens (2DRs) has been confirmed in clinical trials8 and they have been recently incorporated in major guidelines.1–3 Recent analyses from EuroSIDA and RESPOND cohorts suggest that virological, immunological and clinical outcomes in people with HIV (PWH) on 2DRs are similar to PWH on 3DRs.9,10 Knowing the characteristics associated with 2DRs in real life will help to better define the profile of patients who are satisfactorily taking them. We aimed to assess the clinical characteristics associated with the use of 2DRs and the factors associated with specific antiretrovirals in 2DRs in a large single-centre HIV cohort.

Methods

Participants

This retrospective analysis used registered data from all adult PWH prospectively followed at the Hospital Clinic of Barcelona (Spain) who were receiving combination ART with a 3DR or a 2DR in January 2020, when the administrative database was locked for the purpose of this analysis. PWH with last clinical visit before January 2019 were considered lost to follow-up and excluded from the analysis. In our centre, PWH data are routinely registered into a clinical history database approved by the Local Institutional Review Board that includes detailed information on demographics, HIV characteristics and AIDS events, antiretroviral prescription, virological failure and genotypic resistance testing and laboratory results.

Ethics

According to current Spanish regulations, the study was classified as a post-authorization study with a non-prospective design by the Spanish Agency for Medicine and Health Products;11 it was approved by the Local Institutional Review Board and informed consent was not required.

Statistical methods

Summary statistics of quantitative and qualitative variables were reported using mean and SD or median and IQR, using frequency and percentage, respectively. Comparisons between groups were based on t-test, Wilcoxon rank sum, chi-squared and Fisher’s exact tests. First, we described the characteristics of all PWH and compared them in those receiving 3DRs versus 2DRs. We then compared the patients with missing values with those with valid values for all variables in order to identify whether there was a pattern in whether values were missing. Factors associated with the probability of receiving a 2DR relative to a 3DR were assessed using a logistic regression model, controlling for age, sex and also for the characteristic capturing the mechanism of missing values (year of HIV positivity). Considering clinical judgement and statistical criteria (P value < 0.1), variables were selected for the final model in a stepwise fashion. Classification tables and ROC curves were generated by 10-fold cross-validation to avoid an over-optimistic estimate of predictive performance.

Because the drugs more commonly used in 2DRs in this cohort were PIs and integrase inhibitors, the same methodology was applied to identify factors associated with the prescription of integrase inhibitor-based regimens or PI-based regimens among PWH receiving 2DRs. We primarily report risk factors considering two non-exclusive categories: integrase inhibitor-containing versus -non-containing regimens, and PI-containing versus -non-containing regimens, but we explored the factors associated with 2DRs based on both PIs and integrase inhibitors to see whether this group represented a special population. The statistical package used was Stata Release 15 (StataCorp 2017, College Station, TX, USA).

Results

Population characteristics of 2DRs versus 3DRs

There were 3895 PWH included in this analysis: 3432 (88%) receiving 3DRs and 463 (12%) receiving 2DRs. The populations were well distinguished according to demographics. PWH on 2DRs were significantly older [50 years (SD = 11), n = 463 versus 44 years (SD = 11), n = 3432; P < 0.0001] and more commonly female [n = 98 (21%) versus n = 543 (16%); P = 0.0036].

Differences were also evident regarding HIV characteristics and AIDS events, virological failure and genotypic resistance testing. PWH receiving 2DRs had been more commonly infected through injecting drug use [n = 93 (21%) versus n = 385 (12%); P < 0.0001], longer diagnosed with HIV [23 years (IQR 14–28), n = 458 versus 13 years (IQR 8–21), n = 3322; P < 0.0001], more commonly diagnosed with AIDS events [n = 139 (30%) versus n = 502 (15%); P < 0.0001], less commonly on their initial antiretroviral regimen [n = 11 (2%) versus n = 596 (17%); P < 0.0001], longer exposed to ART [17 years (IQR 8–22), n = 463 versus 8 years (IQR 4–15), n = 3432; P < 0.0001], longer on their current regimen [3 years (IQR 2–3), n = 463 versus 2 years (IQR 1–3), n = 3432; P < 0.0001], more commonly had prior virological failure [n = 230 (49%) versus n = 744 (22%); P < 0.0001] and resistance mutations detected [n = 261 (56%) versus n = 1161 (34%); P < 0.0001] but also more commonly had undetectable plasma HIV RNA at the time of initiation of their current regimen [n = 404 (87%) versus n = 2743 (81%); P = 0.0006].

Differences were also found in several laboratory parameters. PWH on 2DRs had higher metabolic parameters at the time of initiation of their current drug regimen: total cholesterol [192 mg/dL (IQR 166–223), n = 449 versus 177 mg/dL (IQR 150–240), n = 3251; P < 0.0001], LDL cholesterol [117 mg/dL (IQR 94–143), n = 444 versus 108 mg/dL (IQR 87–131), n = 3251; P < 0.0001], HDL cholesterol [43 mg/dL (IQR 36–53), n = 449 versus 42 mg/dL (IQR 35–51), n = 3248; P = 0.0241], triglycerides [134 mg/dL (IQR 96–203), n = 449 versus 108 mg/dL (IQR 77–157), n = 3251; P < 0.0001] and glucose [95 mg/dL (IQR 88–105), n = 449 versus 93 mg/dL (IQR 85–101), n = 3253; P = 0.0001]. In contrast, PWH on 2DRs had significantly lower haemoglobin at the time of initiation of their current drug regimen [147 g/L (IQR 137–154), n = 449 versus 148 g/L (IQR 139–157), n = 325; P = 0.0015].

Three thousand three hundred and seventy-five (86.6%) PWH with no missing data contributed to the multivariate analysis of factors associated with the use of 2DRs. In the final model adjusted for age, gender and year of HIV diagnosis, the following independent factors were identified: two or more previous virological failures (adjusted OR 1.69; 95% CI: 1.24–2.31; P = 0.0039), previous resistance mutations (adjusted OR 2.15; 95% CI: 1.70–2.72; P < 0.0001), previous AIDS diagnosis (adjusted OR 1.34; 95% CI: 1.03–1.74; P = 0.0274), longer time on current regimen (adjusted OR 1.14 per year increase; 95% CI: 1.07–1.21; P = 0.0001), higher total cholesterol (adjusted OR 1.32 per 50 unit increase; 95% CI: 1.14–1.53; P = 0.0003) or triglycerides [adjusted OR 1.34 per natural log-transformed (Ln) triglycerides unit increase; 95% CI: 1.08–1.66; P = 0.0074] at the initiation of the current regimen and lower haemoglobin at the initiation of the current regimen (adjusted OR 0.18 per Ln haemoglobin unit increase; 95% CI: 0.06–0.51; P = 0.0013).

Types of 2DRs

There were eight types of regimens identified among the 463 (12%) PWH receiving 2DRs in the cohort. The most common ones were one integrase inhibitor plus one NRTI (n = 112), followed by one integrase inhibitor plus one PI (n = 109), one PI plus one NRTI (n = 82), and one integrase inhibitor plus one NNRTI (n = 80). Other less common 2DRs consisted of one PI plus one NNRTI (n = 49), one PI plus maraviroc (n = 24), one integrase inhibitor plus maraviroc (n = 5), and two NRTIs (n = 2). Dolutegravir (n = 199; 65%) ranked first among integrase inhibitors, followed by raltegravir (n = 107; 35%). Darunavir (n = 240; 91%) was the most common among PIs, followed by atazanavir (n = 16, 6%) and lopinavir (n = 8; 3%). Lamivudine (n = 190; 96%) represented the overwhelming majority of NRTIs, followed by tenofovir (n = 8; 4%). Rilpivirine (n = 74; 57%) was the most common NNRTI, followed by etravirine (n = 49; 38%) and nevirapine (n = 6; 5%). The vast majority of patients (n = 452; 98%) were not on their first antiretroviral regimen.

Factors associated with integrase inhibitor-based 2DRs

Among PWH on 2DRs without missing data, there were 291 PWH prescribed a regimen containing integrase inhibitors and 150 prescribed a regimen not containing integrase inhibitors contributing to this analysis. Lower HDL cholesterol at the initiation of 2DR and longer regimen duration were identified as independent factors associated with an integrase inhibitor in the 2DR (Table 1).

Table 1.
Open in new tab

Univariate and multivariate analysis of factors associated with the inclusion of an integrase inhibitor among patients prescribed a 2DR, adjusted for age at the initiation of two-drug therapy and gender

FactorUnivariate analysis
Multivariate analysis
Adjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.60 (0.29–1.24)0.1680—
Years since HIV diagnosis0.97 (0.94–1.01)0.1608—
Prior AIDS1.58 (0.80–3.14)0.1881—
Years on ART1.02 (0.98–1.07)0.3089—
Years on 2DR1.26 (1.01–1.57)0.03871.26 (1.00–1.58)0.0459
Previous blipsa0.75 (0.43–1.33)0.3275—
Previous virological failure1.63 (0.90–2.93)0.1048—
Prior genotypic resistance mutations0.86 (0.50–1.47)0.5772—
CD4 cellsb,c0.98 (0.94–1.02)0.3233—
CD4/CD8 ratioa,c0.89 (0.51–1.54)0.6789—
Plasma HIV-1 RNA <50 copies/mLc0.86 (0.38–1.96)0.7205—
Total cholesterolb,c0.68 (0.49–0.94)0.0199—
LDL cholesterolb,c0.69 (0.47–1.03)0.0685—
HDL cholesterolb,c0.27 (0.10–0.71)0.00770.27 (0.10–0.73)0.0092
Ln triglyceridesa,c1.23 (0.75–2.02)0.4061—
ASTb,c1.59 (0.55–4.64)0.3938—
Creatininea,c1.17 (0.53–2.62)0.6976—
Ln haemoglobina,c0.26 (0.02–3.73)0.3244—
Ln glucosea,c0.75 (0.14–4.08)0.7417—
FactorUnivariate analysis
Multivariate analysis
Adjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.60 (0.29–1.24)0.1680—
Years since HIV diagnosis0.97 (0.94–1.01)0.1608—
Prior AIDS1.58 (0.80–3.14)0.1881—
Years on ART1.02 (0.98–1.07)0.3089—
Years on 2DR1.26 (1.01–1.57)0.03871.26 (1.00–1.58)0.0459
Previous blipsa0.75 (0.43–1.33)0.3275—
Previous virological failure1.63 (0.90–2.93)0.1048—
Prior genotypic resistance mutations0.86 (0.50–1.47)0.5772—
CD4 cellsb,c0.98 (0.94–1.02)0.3233—
CD4/CD8 ratioa,c0.89 (0.51–1.54)0.6789—
Plasma HIV-1 RNA <50 copies/mLc0.86 (0.38–1.96)0.7205—
Total cholesterolb,c0.68 (0.49–0.94)0.0199—
LDL cholesterolb,c0.69 (0.47–1.03)0.0685—
HDL cholesterolb,c0.27 (0.10–0.71)0.00770.27 (0.10–0.73)0.0092
Ln triglyceridesa,c1.23 (0.75–2.02)0.4061—
ASTb,c1.59 (0.55–4.64)0.3938—
Creatininea,c1.17 (0.53–2.62)0.6976—
Ln haemoglobina,c0.26 (0.02–3.73)0.3244—
Ln glucosea,c0.75 (0.14–4.08)0.7417—
a

Adjusted OR per unit increase.

b

Adjusted OR per 50 unit increase.

c

Laboratory values at the initiation of two-drug therapy.

Table 1.
Open in new tab

Univariate and multivariate analysis of factors associated with the inclusion of an integrase inhibitor among patients prescribed a 2DR, adjusted for age at the initiation of two-drug therapy and gender

FactorUnivariate analysis
Multivariate analysis
Adjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.60 (0.29–1.24)0.1680—
Years since HIV diagnosis0.97 (0.94–1.01)0.1608—
Prior AIDS1.58 (0.80–3.14)0.1881—
Years on ART1.02 (0.98–1.07)0.3089—
Years on 2DR1.26 (1.01–1.57)0.03871.26 (1.00–1.58)0.0459
Previous blipsa0.75 (0.43–1.33)0.3275—
Previous virological failure1.63 (0.90–2.93)0.1048—
Prior genotypic resistance mutations0.86 (0.50–1.47)0.5772—
CD4 cellsb,c0.98 (0.94–1.02)0.3233—
CD4/CD8 ratioa,c0.89 (0.51–1.54)0.6789—
Plasma HIV-1 RNA <50 copies/mLc0.86 (0.38–1.96)0.7205—
Total cholesterolb,c0.68 (0.49–0.94)0.0199—
LDL cholesterolb,c0.69 (0.47–1.03)0.0685—
HDL cholesterolb,c0.27 (0.10–0.71)0.00770.27 (0.10–0.73)0.0092
Ln triglyceridesa,c1.23 (0.75–2.02)0.4061—
ASTb,c1.59 (0.55–4.64)0.3938—
Creatininea,c1.17 (0.53–2.62)0.6976—
Ln haemoglobina,c0.26 (0.02–3.73)0.3244—
Ln glucosea,c0.75 (0.14–4.08)0.7417—
FactorUnivariate analysis
Multivariate analysis
Adjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.60 (0.29–1.24)0.1680—
Years since HIV diagnosis0.97 (0.94–1.01)0.1608—
Prior AIDS1.58 (0.80–3.14)0.1881—
Years on ART1.02 (0.98–1.07)0.3089—
Years on 2DR1.26 (1.01–1.57)0.03871.26 (1.00–1.58)0.0459
Previous blipsa0.75 (0.43–1.33)0.3275—
Previous virological failure1.63 (0.90–2.93)0.1048—
Prior genotypic resistance mutations0.86 (0.50–1.47)0.5772—
CD4 cellsb,c0.98 (0.94–1.02)0.3233—
CD4/CD8 ratioa,c0.89 (0.51–1.54)0.6789—
Plasma HIV-1 RNA <50 copies/mLc0.86 (0.38–1.96)0.7205—
Total cholesterolb,c0.68 (0.49–0.94)0.0199—
LDL cholesterolb,c0.69 (0.47–1.03)0.0685—
HDL cholesterolb,c0.27 (0.10–0.71)0.00770.27 (0.10–0.73)0.0092
Ln triglyceridesa,c1.23 (0.75–2.02)0.4061—
ASTb,c1.59 (0.55–4.64)0.3938—
Creatininea,c1.17 (0.53–2.62)0.6976—
Ln haemoglobina,c0.26 (0.02–3.73)0.3244—
Ln glucosea,c0.75 (0.14–4.08)0.7417—
a

Adjusted OR per unit increase.

b

Adjusted OR per 50 unit increase.

c

Laboratory values at the initiation of two-drug therapy.

Factors associated with PI-based 2DRs

Among patients on 2DRs without missing data, there were 251 PWH prescribed a regimen containing PIs and 188 prescribed a regimen not containing PIs contributing to this analysis. Prior AIDS, prior genotypic resistance mutations, lower CD4/CD8 ratio and shorter regimen duration were identified as independent factors associated with a PI in the 2DR (Table 2). Having previous mutations, previous virological failure and AIDS events were also factors identified when we assessed the population receiving 2DRs based on both PIs and integrase inhibitors.

Table 2.
Open in new tab

Univariate and multivariate analysis of factors associated with the inclusion of a PI among patients prescribed a 2DR, adjusted for age at the initiation of two-drug therapy and gender

Univariate analysis
Multivariate analysis
FactorAdjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.63 (0.37–1.08)0.0947—
Years since HIV diagnosis0.94 (0.91–0.97)0.0001—
Prior AIDS2.16 (1.30–3.56)0.00271.80 (1.01–3.20)0.0474
Years on ART1.04 (1.01–1.08)0.0190—
Years on 2DR0.82 (0.70–0.96)0.01540.78 (0.66–0.93)0.0051
Previous blipsa1.29 (1.05–1.59)0.0160—
Previous virological failure1.46 (1.22–1.76)0.0001—
Prior genotypic resistance mutations5.35 (3.34–8.58)<0.00016.03 (3.61–10.08)<0.0001
CD4 cellsb,c0.97 (0.94–1.00)0.0471—
CD4/CD8 ratioa,c0.41 (0.25–0.67)0.00040.51 (0.29–0.89)0.0168
Plasma HIV-1 RNA <50 copies/mLc0.37 (0.18–0.76)0.0063—
Total cholesterolb,c1.34 (1.03–1.74)0.0266—
LDL cholesterolb,c1.30 (0.96–1.77)0.0851—
HDL cholesterolb,c1.47 (0.66–3.27)0.3500—
Ln triglyceridesa,c1.45 (0.98–2.14)0.0654—
ASTb,c0.80 (0.41–1.55)0.5093—
Creatininea,c0.66 (0.36–1.19)0.1671—
Ln haemoglobina,c0.68 (0.11–4.04)0.6694—
Ln glucosea,c3.24 (0.84–12.57)0.0886—
Univariate analysis
Multivariate analysis
FactorAdjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.63 (0.37–1.08)0.0947—
Years since HIV diagnosis0.94 (0.91–0.97)0.0001—
Prior AIDS2.16 (1.30–3.56)0.00271.80 (1.01–3.20)0.0474
Years on ART1.04 (1.01–1.08)0.0190—
Years on 2DR0.82 (0.70–0.96)0.01540.78 (0.66–0.93)0.0051
Previous blipsa1.29 (1.05–1.59)0.0160—
Previous virological failure1.46 (1.22–1.76)0.0001—
Prior genotypic resistance mutations5.35 (3.34–8.58)<0.00016.03 (3.61–10.08)<0.0001
CD4 cellsb,c0.97 (0.94–1.00)0.0471—
CD4/CD8 ratioa,c0.41 (0.25–0.67)0.00040.51 (0.29–0.89)0.0168
Plasma HIV-1 RNA <50 copies/mLc0.37 (0.18–0.76)0.0063—
Total cholesterolb,c1.34 (1.03–1.74)0.0266—
LDL cholesterolb,c1.30 (0.96–1.77)0.0851—
HDL cholesterolb,c1.47 (0.66–3.27)0.3500—
Ln triglyceridesa,c1.45 (0.98–2.14)0.0654—
ASTb,c0.80 (0.41–1.55)0.5093—
Creatininea,c0.66 (0.36–1.19)0.1671—
Ln haemoglobina,c0.68 (0.11–4.04)0.6694—
Ln glucosea,c3.24 (0.84–12.57)0.0886—
a

Adjusted OR per unit increase.

b

Adjusted OR per 50 unit increase.

c

Laboratory values at the initiation of two-drug therapy.

Table 2.
Open in new tab

Univariate and multivariate analysis of factors associated with the inclusion of a PI among patients prescribed a 2DR, adjusted for age at the initiation of two-drug therapy and gender

Univariate analysis
Multivariate analysis
FactorAdjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.63 (0.37–1.08)0.0947—
Years since HIV diagnosis0.94 (0.91–0.97)0.0001—
Prior AIDS2.16 (1.30–3.56)0.00271.80 (1.01–3.20)0.0474
Years on ART1.04 (1.01–1.08)0.0190—
Years on 2DR0.82 (0.70–0.96)0.01540.78 (0.66–0.93)0.0051
Previous blipsa1.29 (1.05–1.59)0.0160—
Previous virological failure1.46 (1.22–1.76)0.0001—
Prior genotypic resistance mutations5.35 (3.34–8.58)<0.00016.03 (3.61–10.08)<0.0001
CD4 cellsb,c0.97 (0.94–1.00)0.0471—
CD4/CD8 ratioa,c0.41 (0.25–0.67)0.00040.51 (0.29–0.89)0.0168
Plasma HIV-1 RNA <50 copies/mLc0.37 (0.18–0.76)0.0063—
Total cholesterolb,c1.34 (1.03–1.74)0.0266—
LDL cholesterolb,c1.30 (0.96–1.77)0.0851—
HDL cholesterolb,c1.47 (0.66–3.27)0.3500—
Ln triglyceridesa,c1.45 (0.98–2.14)0.0654—
ASTb,c0.80 (0.41–1.55)0.5093—
Creatininea,c0.66 (0.36–1.19)0.1671—
Ln haemoglobina,c0.68 (0.11–4.04)0.6694—
Ln glucosea,c3.24 (0.84–12.57)0.0886—
Univariate analysis
Multivariate analysis
FactorAdjusted OR (95% CI)P valueAdjusted OR (95% CI)P value
Presumed route of HIV infection MSM0.63 (0.37–1.08)0.0947—
Years since HIV diagnosis0.94 (0.91–0.97)0.0001—
Prior AIDS2.16 (1.30–3.56)0.00271.80 (1.01–3.20)0.0474
Years on ART1.04 (1.01–1.08)0.0190—
Years on 2DR0.82 (0.70–0.96)0.01540.78 (0.66–0.93)0.0051
Previous blipsa1.29 (1.05–1.59)0.0160—
Previous virological failure1.46 (1.22–1.76)0.0001—
Prior genotypic resistance mutations5.35 (3.34–8.58)<0.00016.03 (3.61–10.08)<0.0001
CD4 cellsb,c0.97 (0.94–1.00)0.0471—
CD4/CD8 ratioa,c0.41 (0.25–0.67)0.00040.51 (0.29–0.89)0.0168
Plasma HIV-1 RNA <50 copies/mLc0.37 (0.18–0.76)0.0063—
Total cholesterolb,c1.34 (1.03–1.74)0.0266—
LDL cholesterolb,c1.30 (0.96–1.77)0.0851—
HDL cholesterolb,c1.47 (0.66–3.27)0.3500—
Ln triglyceridesa,c1.45 (0.98–2.14)0.0654—
ASTb,c0.80 (0.41–1.55)0.5093—
Creatininea,c0.66 (0.36–1.19)0.1671—
Ln haemoglobina,c0.68 (0.11–4.04)0.6694—
Ln glucosea,c3.24 (0.84–12.57)0.0886—
a

Adjusted OR per unit increase.

b

Adjusted OR per 50 unit increase.

c

Laboratory values at the initiation of two-drug therapy.

Discussion

In our cohort, 1 out of 10 PWH receiving combination ART had 2DRs. Relative to 3DRs, PWH on 2DRs were older and more virologically suppressed, with a longer history of HIV infection and exposure to ART, prior AIDS, prior virological failure and genotypic resistance mutations, higher plasma lipids and lower haemoglobin. These findings are in agreement with previous reports showing that 2DRs have been preferentially addressed to older PWH with a good HIV status but limited therapeutic options because of resistance, cumulative antiretroviral toxicity or concomitant comorbidities.5,9,12,13

The vast majority of 2DRs in this cohort included an integrase inhibitor, a PI or both. We identified different profiles of factors associated with having either an integrase inhibitor or a PI included in the 2DR. Lower baseline HDL cholesterol and longer duration were independent factors associated with having an integrase inhibitor-containing 2DR. Prior AIDS, prior genotypic resistance mutations, lower CD4/CD8 ratio and shorter duration were independent factors associated with having a PI-containing 2DR. The differences in the duration of the 2DR prescription may be related, at least in part, to the better tolerability and lower risk of drug–drug interactions of integrase inhibitors compared with PIs. Use of PIs was associated with a worse HIV status and prior resistance mutations. Use of integrase inhibitors was associated with lower baseline HDL cholesterol. Low HDL cholesterol is a criterion of metabolic syndrome and a marker of insulin resistance and cardiovascular risk.14 We also explored the factors associated with 2DRs based on both PIs and integrase inhibitors, which included having previous mutations, previous virological failure and AIDS events. Therefore, it was the presence of prior virological failure with resistance mutations that was related to inclusion of PIs with or without integrase inhibitors.

In this large single-centre HIV cohort, 2DRs had been tailored according to the clinical characteristics of PWH and usually contained PIs or/and integrase inhibitors. A worse cardiometabolic status or more archived resistance were key factors associated with the inclusion of integrase inhibitors or PIs, respectively, in 2DRs.

Funding

The study was funded by Instituto de Salud Carlos III (grant numbers PI16/01085 and PI20/00869).

Transparency declarations

A. Gonzalez-Cordon, A. Inciarte, M. Martinez-Rebollar, M. Laguno, J. Ambrosioni, B. Torres, J. Mallolas, J. L. Blanco, J. M. Miro and E. Martinez have received honoraria for lectures or advisory boards and their institution has received research grants from Gilead, Janssen, MSD and ViiV. E. de Lazzari, A. Ugarte and L. de la Mora: none to declare.

Author contributions

E. Martinez designed the study. E. de Lazzari undertook the statistical analyses. All authors were involved in the interpretation of data. E. Martinez and E. de Lazzari drafted the manuscript. All authors critically reviewed and subsequently approved the final version.

References

1

Panel de Expertos de GeSIDA y Plan Nacional sobre el SIDA. Documento de consenso de GeSIDA/Plan Nacional sobre el SIDA respecto al tratamiento antirretroviral en adultos infectados por el virus de la inmunodeficiencia humana (actualización

2020
). http://gesida-seimc.org/wp-content/uploads/2020/07/TAR_GUIA_GESIDA_2020_COMPLETA_Julio.pdf.

2

European AIDS Clinical Society. Guidelines (version 10.1 October

2020
). https://www.eacsociety.org/files/guidelines-10.1_5.pdf.

3

Saag
MS
,
Gandhi
RT
,
Hoy
JF
 et al.
Antiretroviral drugs for treatment and prevention of HIV infection in adults: 2020 recommendations of the International Antiviral Society–USA Panel
.
JAMA
2020
;
324
:
1651
–
69
.

Google Scholar

Crossref
Search ADS
PubMed

 
4

Guaraldi
G
,
Menozzi
M
,
Zona
S
 et al.
Impact of polypharmacy on antiretroviral prescription in people living with HIV
.
J Antimicrob Chemother
2017
;
72
:
511
–
4
.

Google Scholar

Crossref
Search ADS
PubMed

 
5

Martinez
E
,
Negredo
E
,
Knobel
H
 et al.
Factors associated with the number of drugs in darunavir/cobicistat regimens
.
J Antimicrob Chemother
2020
;
75
:
208
–
14
.

Google Scholar

Crossref
Search ADS
PubMed

 
6

Arribas
JR
,
Girard
PM
,
Paton
N
 et al.
Efficacy of protease inhibitor monotherapy vs. triple therapy: meta-analysis of data from 2303 patients in 13 randomized trials
.
HIV Med
2016
;
17
:
358
–
67
.

Google Scholar

Crossref
Search ADS
PubMed

 
7

Blanco
JL
,
Marcelin
AG
,
Katlama
C
 et al.
Dolutegravir resistance mutations: lessons from monotherapy studies
.
Curr Opin Infect Dis
2018
;
31
:
237
–
45
.

Google Scholar

Crossref
Search ADS
PubMed

 
8

Pisaturo
M
,
Onorato
L
,
Russo
A
 et al.
Risk of failure in dual therapy versus triple therapy in naïve HIV patients: a systematic review and meta-analysis
.
Clin Microbiol Infect
2021
;
27
:
28
–
35
.

Google Scholar

Crossref
Search ADS
PubMed

 
9

Neesgaard
B
,
Pelchen-Matthews
A
,
Ryom
L
 et al.
Uptake and effectiveness of two-drug compared with three-drug antiretroviral regimens among HIV-positive individuals in Europe
.
AIDS
2019
;
33
:
2013
–
24
.

Google Scholar

Crossref
Search ADS
PubMed

 
10

Greenberg
L
,
Ryom
L
,
Neesgaard
B
 et al.
Clinical outcomes of two-drug regimens vs. three-drug regimens in antiretroviral treatment-experienced people living with HIV
.
Clin Infect Dis
2020
; ciaa1878. https://doi:10.1093/cid/ciaa1878.

Google Scholar

OpenURL Placeholder Text

 
11

BOE. Orden SAS/3470/2009, de 16 de diciembre, por la que se publican las directrices sobre studios posautorizacion de tipo observacional para medicamentos de uso humano. https://www.boe.es/boe/dias/2009/12/25/pdfs/BOE-A-2009-20817.pdf.

12

Monteiro
P
,
Perez
I
,
Laguno
M
 et al.
Dual therapy with etravirine plus raltegravir for virologically suppressed HIV-infected patients: a pilot study
.
J Antimicrob Chemother
2014
;
69
:
742
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
13

Nozza
S
,
Malagoli
A
,
Maia
L
 et al.
Antiretroviral therapy in geriatric HIV patients: the GEPPO cohort study
.
J Antimicrob Chemother
2017
;
72
:
2879
–
86
.

Google Scholar

Crossref
Search ADS
PubMed

 
14

van Wijk
JP
,
Cabezas
MC.
Hypertriglyceridemia, metabolic syndrome, and cardiovascular disease in HIV-infected patients: effects of antiretroviral therapy and adipose tissue distribution
.
Int J Vasc Med
2012
;
2012
:
201027
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
© The Author(s) 2021. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: [email protected].
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://dbpia.nl.go.kr/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Topic:
Issue Section:
ORIGINAL RESEARCH
Download all slides