Insights from systematic reviews (2019–2024) and drug interaction database analysis in people with HIV and comorbidities

Abstract

Background

This study investigates drug-drug interactions (DDIs) in people with HIV (PWH) receiving antiretroviral therapy (ART) with comorbidities. We focus on pharmacological factors and evaluate DDI notifications from online databases, emphasizing the clinical relevance of pharmacokinetic, pharmacodynamic, and pharmacogenomic variations.

Methods

A comprehensive literature search was conducted using PubMed, ScienceDirect, Google Scholar, and the Cochrane Library for studies published between January 2019 and September 2024. Newly identified DDI evidence was analyzed by comparing DrugBank, Drugs.com, and the Liverpool HIV databases.

Results

Eleven studies met the inclusion criteria. The findings of these studies showed the critical importance of considering DDIs in PWH with tuberculosis (TB), malaria, and pulmonary hypertension. Severe adverse drug reactions associated with ARTs, including efavirenz, darunavir, nevirapine, and atazanavir-ritonavir, especially when combined with treatments for TB and malaria. Key interactions included reduced drug levels from rifampicin and QT prolongation from artemether-lumefantrine. Pharmacogenomic factors, such as CYP2B6 slow metabolism during pregnancy, influenced outcomes. Database discrepancies were noted, especially for riociguat interactions and ritonavir through inhibition of P-gp or OATP1B1 functions.

Conclusions

DDIs in PWH receiving ART with comorbidities have highlighted the crucial need for personalized treatment. Incorporating pharmacokinetic, pharmacodynamic, and pharmacogenomic factors is essential for optimizing therapy outcomes.

Keywords

drug-drug interactions antiretroviral therapy people with HIV and adverse drug reactions

Introduction

The management of human immunodeficiency virus (HIV) has significantly improved with the introduction of antiretroviral therapy (ART), leading to better long-term outcomes and quality of life for people with HIV (PWH).^1,2 However, as ART regimens become increasingly complex, particularly in individuals with comorbid conditions such as tuberculosis (TB), malaria, cardiovascular disease, and mental health disorders, the risk of severe drug–drug interactions (DDIs) also increases.^3–6 Epidemiological evidence suggests that coadministration of ART with other medications may heighten the risk of toxicity, therapeutic failure, and adverse drug reactions (ADRs).^3,6–9 For instance, efavirenz has been shown to reduce plasma concentrations of atovaquone/proguanil by 43%–75%, potentially compromising malaria prophylaxis.⁸ Protease inhibitors such as atazanavir/cobicistat and darunavir/cobicistat have been reported to significantly increase rosuvastatin levels, with corresponding risks of rhabdomyolysis and severe myopathy.⁹ DDIs are particularly prevalent among vulnerable populations, including older adults, pregnant individuals, and those with multiple comorbidities,^7,10,11 further complicating HIV treatment strategies.

Many ART-related DDIs involve pharmacokinetic and pharmacodynamic mechanisms, particularly cytochrome P450 enzymes such as CYP3A4 and CYP2C9.⁶ DDIs contribute to various ADRs.^3,6,10–12 A deeper understanding of the mechanistic pathways of ART-related DDIs is essential for optimizing treatment outcomes and minimizing harm. Despite an expanding body of literature on ART-associated DDIs, most prior reviews (1995–2022) had focused narrowly on food interactions, herbal products, and single-drug analyses.⁹ There remains a lack of comprehensive evaluation of the full spectrum of ART-related DDIs, particularly in PWH with co-existing medical conditions.^9,12–14 Moreover, mechanistic factors such as drug metabolizing enzyme induction/inhibition, and transporter modulation are frequently underexplored. Another underrepresented area is the role of pharmacogenomic variability in determining individual DDI outcomes.¹⁵ Considering these factors in HIV management, including disease burden, polypharmacy, and genetic variability, can improve treatment safety and efficacy, highlighting the need for a comprehensive approach to HIV treatment. In addition, clinical guidelines and recommendations for HIV treatment have been recently updated.¹⁶

This study aims to synthesize and evaluate recent evidence on DDI involving ART, focusing on clinical studies published between January 2019 and September 2024 in the four high-impact databases (the Cochrane Library, PubMed, Google Scholar, and ScienceDirect). This literature review followed systematic review principles based on PRISMA 2020 guidelines. This timeframe reflects recent advancements in ART and evolving drug interaction patterns associated with emerging treatment regimens. In addition to summarizing findings related to pharmacokinetics, pharmacodynamics, and pharmacogenomics, we compared the included studies with three widely used drug interaction resources: DrugBank, Drugs.com, and the Liverpool HIV interaction database to note new information on DDI evidence. These platforms are strengths: DrugBank for detailed pharmacological insights, Drugs.com for broad clinical accessibility, and Liverpool HIV for HIV-specific expertise. The database comparison was used to assess the consistency and coverage of DDI. With its comprehensive approach, this study identified discrepancies in DDI classifications, underlying mechanisms, and severity ratings by integrating data from peer-reviewed literature with curated interaction databases. Our finding has practical implications for managing DDIs in HIV care, equipping healthcare providers with the knowledge to inform personalized strategies and improve patient safety and therapeutic outcomes, thereby instilling confidence in their practice.

Methods

PRISMA compliance

The PRISMA 2020 guidelines were used to conduct this systematic review.¹⁷ The completed PRISMA checklist is available in Supplemental Table S1.

Review protocol

The protocol was not registered in PROSPERO or any other database due to the retrospective nature of the review process. This decision was made to avoid potential bias in the selection of studies, as the review was based on existing literature and did not involve any prospective data collection.

Data searching strategy

Our comprehensive literature search, conducted on 12 September 2024 across four databases (PubMed, Cochrane Library, Google Scholar, and ScienceDirect), was meticulously planned and executed. We used Medical Subject Headings (MeSH) and relevant keywords, including HIV/AIDS, anti-HIV, drug interaction, antiretroviral therapy, treatment, and clinical pharmacology. The search was limited to English-language studies published between 1 January 2019 and 12 September 2024. The search strategy for each database is shown in Supplemental Table S2.

Study selection and data extraction

Six reviewers (PB, LR, PT, PR, NC, CB) were involved in the study selection and data extraction process. Titles and abstracts were independently screened for eligibility by reviewers based on predefined criteria. Full-text screening was conducted for studies deemed potentially relevant. Any disagreements were resolved through discussion and consensus among the reviewers, ensuring the reliability of our process.

Inclusion criteria were:

Peer-reviewed clinical studies involving HIV-positive individuals with comorbidities

Studies reporting DDIs involving ART

Studies reporting data on: sample size, follow-up duration, pharmacokinetic profiles (C_max, T_max, AUC), therapeutic efficacy, and/or ADRs

Exclusion criteria included:

Studies not involving HIV-positive populations

Review articles (conference abstracts, systematic, narrative, or meta-analyses)

Preclinical studies (in vitro or in vivo animal models)

Articles with incomplete, unclear, or insufficient data

Data from included studies were extracted using a standardized form in Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). Extracted data fields included: ART regimen design, population characteristics, co-medications, reported DDIs, pharmacokinetic/pharmacodynamic/pharmacogenetic findings, ADRs, and drug resistance.

Research questions

The following research questions were addressed:

What is the frequency of drug–ART interactions in PWH, with or without comorbidities, and how are these interactions classified as mild, moderate, or severe according to the Common Terminology Criteria for Adverse Events (CTCAE)?

Which drugs interact with ART, resulting in pharmacokinetic, pharmacodynamic, or pharmacogenomic changes that impact clinical outcomes?

What are the mechanisms of worsened ADRs due to drug-ART interactions, as outlined in DrugBank, Drugs.com, and Liverpool HIV interactions?

Quality assessment and risk of bias

The quality of included studies was assessed using the Cochrane Collaboration’s Risk of Bias Tool, as described in the Cochrane Handbook for Systematic Reviews of Interventions.¹⁸ The following domains were evaluated: sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other sources of bias. Each domain was classified as having low, high, or unclear risk of bias. A summary of risk assessments is provided in Supplemental Table S3.

Outcome analysis

Outcomes were categorized based on the frequency and severity of DDIs (mild, moderate, or severe) from clinical data, with visualization via a Sankey diagram to illustrate the relationship between drug classes and interaction severity. Clinical outcomes were evaluated regarding pharmacokinetic and pharmacodynamic changes, pharmacogenomic associations, ADRs, and instances of drug resistance.

Drug interaction database comparison

To strengthen the clinical relevance of the findings, DDIs identified in the literature were cross-referenced with three widely used drug interaction databases as follows:

DrugBank - detailed pharmacological data (accessed 15 October 2024)

Drugs.com - user-friendly clinical interface (accessed 15 October 2024)

Liverpool HIV Interaction Database - HIV-specific interactions (accessed 20 February 2025)

These resources were chosen for their currency, comprehensiveness, and clinical utility. The comparison allowed identification of discrepancies in interaction severity, mechanistic classification, and notification systems. Findings from this cross-comparison informed a more robust pharmacovigilance framework and clinical interpretation of DDIs in HIV care.

Results

Article selection, screening process, and risk of bias

PRISMA flow diagram shown in Figure 1. We identified 70 articles across the four databases. After removing 20 duplicates, 50 titles and abstracts were screened. Of these, 32 articles were selected for full-text review, but six were unavailable despite repeated attempts. Of the 26 reviewed, 15 articles were excluded due to a lack of HIV-positive populations, pharmacokinetic/pharmacodynamic data, ADR reporting, or data clarity. Ultimately, 11 high-quality studies met the inclusion criteria for data extraction and synthesis. Risk of bias was assessed using the Cochrane Collaboration’s tool. Studies varied significantly in blinding, allocation concealment, and outcome reporting (Supplemental Table S3), with some rated high risk and others low, indicating a need for improved methodological rigor in future research.

Figure 1.

The PRISMA flow diagram depicted the process of selecting articles for the systematic review.

Frequency and severity of drug–ART interactions in PWH

Based on a synthesis of 11 studies, the review identified frequent DDIs involving ART as visualized in a Sankey diagram (Figure 2) and summarized in Table 1. Twelve ART agents were associated with varying degrees of interaction severity. Efavirenz exhibited moderate to severe DDIs when administered with tuberculosis treatments such as rifampicin and isoniazid, as well as with the antimalarial combination artemether-lumefantrine. Riociguat demonstrated moderate interactions with abacavir, lamivudine, and dolutegravir, while weaker interactions were observed between riociguat and other agents, including ritonavir, emtricitabine, tenofovir disoproxil fumarate, rilpivirine, and elvitegravir. Clinically significant adverse effects were noted in particular combinations, such as atazanavir-ritonavir with artemether-lumefantrine, and nevirapine with either rifapentine or isoniazid. In contrast, the combination of tenofovir disoproxil fumarate and dolutegravir with islatravir was well tolerated, with no substantial adverse effects reported.

Figure 2.

The frequency and severity of drug-ART interactions observed in PWH.

Table 1.

Outcomes of drug interactions with antiretroviral drugs from the 11 included studies.

Drugs	Antiretroviral drugs (ARTs)	Participants	Clinical outcomes	Pharmacokinetic or pharmacodynamic changes and pharmacogenomics	Adverse drug reactions (ADRs)/Drug resistance	Author and year
Severe drug-drug interactions
Malaria drug:	NNRTIs:	186 participants with uncomplicated malaria caused by P. falciparum and HIV (both male and female, aged over 18 years)	Decreased lumefantrine and desbutyl-lumefantrine (active metabolite) levels, leading to treatment failure and high parasite density in 94 patients (P < 0.008)	Decreased blood levels of lumefantrine and desbutyl-lumefantrine	Treatment failure: 61% of participants had low lumefantrine levels (<200 ng/mL) and 37.7% had undetectable desbutyl-lumefantrine. High parasite density was observed in participants on efavirenz-based ARTs.	Nyangulu et al., 2023¹⁹
Artemether-Lumefantrine	Efavirenz-based ARTs				Potential for drug resistance: Sub-therapeutic levels of lumefantrine may allow for the survival of resistant parasites, potentially leading to the spread of resistance.
TB drug:	PIs:	17 participants (both male and female, aged 18-60 years) with complications of TB	Decreased plasma levels of darunavir and ritonavir, as well as in PBMCs, after treatment with rifampicin for 2 weeks (P < 0.001)	Decreased CTrough and AUC of darunavir and ritonavir in plasma levels and PBMCs after adding rifampicin	Severe ADR: Symptoms included hepatotoxicity (grade 3 and 4 ALT elevation).	Nicolò et al., 2022²¹
Rifampicin (600 -750 mg)	Ritonavir-boosted darunavir (100 mg and 800 mg) and ritonavir (200 mg)		Hepatic toxicity was observed after adding rifampicin when co-treated with darunavir and ritonavir in 12 patients.	No change in dolutegravir Ctrough and AUC in plasma levels and PBMCs when combined with rifampicin	Study cessation: Hepatic toxicity observed after adding rifampicin during co-treatment with ritonavir-boosted darunavir.
	INSTI:
	Dolutegravir regimen (50 mg) in ARTs
TB drug:	NNRTI:	ART containing nevirapine (200 mg), 78 evaluable participants with HIV-TB (all male and female, aged 13–66 years)	Decreased bioavailability of nevirapine when combined with rifapentine/isoniazid at week 4 by approximately 29%	Reduced concentration of nevirapine in plasma (<3000 ng/mL)	Potential loss of virological suppression: Increased viral load >40 copies/ml at 4 to 8 weeks	Podany et al., 2021²²
Rifapentine (600 mg) and isoniazid (300 mg)	ART containing nevirapine (200 mg)			Increased oral clearance (CL/F) of nevirapine by 30% with concomitant 1HP
TB drug:	NNRTI:	HIV-positive pregnant and postpartum women with TB (aged ≥18, gestational age 14–34 weeks, and weighing >35 kg) who are NAT2 and CYP2B6 genotypes	Rapid NAT2 acetylators and CYP2B6 metabolizers, as well as pregnancy, increased isoniazid and efavirenz clearance by 26% and 15%, respectively (P < 0.001).	The AUCs of isoniazid and efavirenz exposures were reduced during pregnancy due to increased clearance.	Potential treatment failure and drug resistance:	Gausi et al., 2020²³
Isoniazid (300 mg)	ART containing efavirenz		Isoniazid decreased efavirenz clearance by 7% in CYP2B6 normal metabolizers and 13% in slow and intermediate metabolizers (P < 0.001).	In slow and intermediate metabolizers of NAT2 and CYP2B6, efavirenz plasma concentrations increased when co-administered with isoniazid.	There is an increased risk of sub-therapeutic drug levels of isoniazid and efavirenz during pregnancy, leading to ineffective treatment, which could allow HIV and TB to progress or develop resistance, especially in CYP2B6 normal metabolizers or NAT2 rapid acetylators.
					Severe ADR:
					Higher risk of efavirenz toxicity in CYP2B6 slow metabolizers when administered with isoniazid, leading to neuropsychiatric symptoms and hepatotoxicity
Malaria drug:	PI:	10 participants with P. falciparum uncomplicated malaria and HIV (all male and female, aged ≥18 years)	Atazanavir-ritonavir regimen in ART resulted in QT prolongation (from 0.4079 to 0.4215s) when combined with artemether-lumefantrine (p = 0.008).	Increased QT interval prolongation was observed with the administration of the atazanavir-ritonavir regimen and artemether-lumefantrine	Severe ADR: Potential cardiotoxicity, highlighted by a significant increase in QT interval	Usman et al., 2022²⁰
Artemether/lumefantrine (80/480 mg)	Atazanavir-ritonavir (300/100 mg) in ART regimen				Common ADRs: Fever, headache, general weakness, and chest pain
					Clinical Monitoring: No significant changes were observed in liver (ALT, AST), renal (creatinine), or hematologic (hemoglobin) parameters
Moderate drug-drug interactions
Pulmonary hypertension drug:	Several drug combinations were used, including:	40 participants with pulmonary arterial hypertension and HIV (male, aged 18–64 years)	Increased riociguat levels were observed when combined with abacavir, dolutegravir, and lamivudine, resulting in higher Cmax and AUC (up to threefold) compared to the following four ART regimens: Elvitegravir/cobicistat/emtricitabine/tenofovir disoproxil > Emtricitabine/rilpivirine/tenofovir disoproxil > Ritonavir-boosted triple regimen > Efavirenz/emtricitabine/tenofovir disoproxil.	Increased Cmax and AUC of riociguat in the ART regimen of abacavir, dolutegravir, and lamivudine	Common ADRs: Headache and Fatigue	DeJesus et al., 2019²⁴
A single oral dose of riociguat (0.5 mg)	2 NRTIs and NNRTI:				Moderate ADR: Stomach cramps related to riociguat
	Emtricitabine (200 mg), tenofovir disoproxil fumarate (300 mg), and efavirenz (600 mg) (ATRIPLA)				Clinical monitoring: No significant changes in vital signs, ECG, or laboratory parameters were observed. No ADRs related to decreased blood pressure or hypotension were reported.
	Emtricitabine (200 mg, rilpivirine (25 mg), and tenofovir disoproxil fumarate (300 mg) (COMPLERA)
	2 NRTIs and INSTI:
	Elvitegravir (150 mg), cobicistat (150 mg), emtricitabine (200 mg), and tenofovir disoproxil fumarate (300 mg) (STRIBILD)
	Abacavir (600 mg), dolutegravir (50 mg), and lamivudine (300 mg) (TRIUMEQ)
	PI:
	Ritonavir-boosted triple regimen.
TB drug:	INSTI and NRTI:	128 participants (both male and female HIV-TB co-infected, aged >18 years)	Reduced dolutegravir and efavirenz plasma levels were observed in patients receiving high-dose rifampicin, but not in those receiving standard-dose.	Decreased dolutegravir and efavirenz blood concentrations were noted.	General tolerability: Among patients with good tolerance	Sekaggya-Wiltshire et al., 2023²⁵
Standard-dose (600 mg) and high-dose (1800 mg) of rifampicin in the TB regimen	Dolutegravir (50 mg) and efavirenz (600 mg)-based ART regimen		The incidence of HIV treatment failure at week 24 was similar between the rifampicin doses (14.9% vs. 14.0%, p = 0.901).		Severe ADR: Some patients experienced grade 3-4 adverse drug reactions.
					Clinical monitoring: Some patients showed increased liver enzyme levels (elevated ALT).
					Sputum conversion at week 8 was higher in the high-dose versus the standard-dose arms.
TB drug:	INSTI:	44 individuals (both male and female, aged 31-43) with TB complications and UGT1A1 genotypes (1/36, 1/1, 1/28, 28/28, and unknown)	Decreased dolutegravir bioavailability when exposed to rifampicin, but therapeutic efficacy is maintained	Reduced AUC of dolutegravir, independent of UGT1A1 gene polymorphisms	Clinical finding: The use of higher dolutegravir dosing regimens in patients receiving high-dose rifampicin may increase the risk of side effects such as gastrointestinal disturbances, insomnia, or other central nervous system effects.	Kengo et al., 2023²⁶
Standard-dose (600 mg) and high-dose (2100 mg) rifampicin included TB regimen	Dolutegravir-based ART regimen (50 mg)
Low drug-drug interactions
Novel NRTI:	NRTI:	12 participants (all males and females, aged 18 to 55 years)	No clinically meaningful change in dolutegravir and tenofovir disoproxil fumarate levels, with no significant changes in HIV suppression	No change in AUC, Cmax, or half-life values	Common ADRs: Headache, abdominal discomfort, and nausea	Rudd et al., 2021²⁷
A single oral dose of islatravir (20 mg)	Tenofovir disoproxil fumarate (300 mg)				Clinical monitoring: No significant changes in lab tests, vital signs, or ECG
	INSTI:
	Dolutegravir (50 mg)
TB drug:	NNRTI:	102 HIV co-infected TB participants (all males and females, aged 27 to 53 years)	Initiating ART after 4 or 8 weeks of anti-TB therapy resulted in no significant differences in the extent of CYP3A4 induction and CYP3A-mediated rifampicin-efavirenz drug interactions	No change in efavirenz plasma concentrations	No new ADRs reported.	Aklillu et al., 2020²⁸
Rifampicin included in TB therapy	Efavirenz (600 mg) in ARTs (lamivudine, zidovudine, or stavudine)
TB drug:	NRTI:	97 HIV co-infected TB participants in a study of CYP2B6 genetic polymorphism (slow, intermediate, and extensive metabolizers) (all males and females, aged 27 to 43 years)	No clinically meaningful change in efavirenz plasma levels when combined with rifampicin-based TB regimen	No change AUC of efavirenz	Severe ADR: Some patients experienced grade 3 or 4 ADR such as anemia and thrombocytopenia.	Atwine et al., 2020²⁹
Rifampicin included in TB drug regimen (150-300 mg)	Efavirenz (600 to 800 mg) in ART regimen				Clinical monitoring: Some patients show an increase in liver enzyme AST and ALT levels

Note. INSTI, integrase strand transfer inhibitor; NNRTIs, non-nucleoside reverse transcriptase inhibitor; NRTI, nucleoside reverse transcriptase inhibitor; PI, protease inhibitor; TB, Tuberculosis.

Impact on pharmacokinetics, pharmacodynamics, and pharmacogenomics

As shown in Table 1, the reviewed literature demonstrated DDIs between ARTs and antimalarial or antituberculosis medications, with important implications for efficacy and safety in HIV co-infected individuals. The efavirenz-based regimen was found to significantly reduce plasma lumefantrine levels, resulting in subtherapeutic concentrations and treatment failure in 61% of patients with malaria co-infection, potentially increasing parasitemia and the risk of resistance development.¹⁹ Atazanavir-ritonavir administered with artemether-lumefantrine was associated with QT interval prolongation, elevating the risk of cardiotoxicity.²⁰ Rifampicin, when combined with ritonavir-boosted darunavir, reduced ritonavir-boosted darunavir exposure.²¹ Similarly, rifapentine and isoniazid coadministration decreased nevirapine bioavailability and increased its clearance, reducing virologic control.²² In pregnant women with the CYP2B6 slow metabolizer genotype, efavirenz plasma concentrations increased when combined with isoniazid, raising the risk of neuropsychiatric and hepatic toxicity.²³ Conversely, women with rapid NAT2 and normal CYP2B6 genotypes showed lower isoniazid and efavirenz plasma levels, potentially leading to subtherapeutic drug exposure during pregnancy.²³

Integrase inhibitors also showed notable interactions. Coadministration of abacavir, dolutegravir, and lamivudine with riociguat increased riociguat plasma levels in patients with HIV-associated pulmonary hypertension, resulting in moderate ADRs such as abdominal discomfort.²⁴ While high-dose rifampicin reduced dolutegravir and efavirenz plasma concentrations, virologic suppression was maintained.²⁵ The decrease of AUC in dolutegravir appeared independent of UGT1A1 polymorphisms; however, higher dolutegravir doses may lead to insomnia and gastrointestinal side effects.²⁶

In contrast, islatravir, a novel nucleoside reverse transcriptase inhibitor, did not significantly alter the pharmacokinetics of dolutegravir or tenofovir disoproxil fumarate and preserved HIV suppression.²⁷ Furthermore, low to moderate doses of rifampicin (150–300 mg) showed minimal interaction with efavirenz, supporting the continued use of current dosing regimens in co-infected populations.^28,29

Mechanistic insights from literature reviews compared with drug interaction databases

Our study further examined the mechanisms underlying ART-related DDIs, drawing on evidence from the included literature and comparing these findings with data from three widely used drug interaction databases: DrugBank, Drugs.com, and the Liverpool HIV interaction checker (Table 2). The predominant mechanisms involved cytochrome P450 (CYP) enzyme induction or inhibition, particularly CYP1A1, CYP3A4, and UGT1A1, and inhibition of drug transporters such as P-gp and BCRP. Literature-based evidence indicated that abacavir, dolutegravir, and lamivudine may inhibit CYP1A1 and P-gp/BCRP, potentially reducing the metabolism of riociguat.²⁴ DrugBank identified dolutegravir as a P-gp inhibitor, while Drugs.com attributed CYP1A1 inhibition to abacavir. In contrast, the Liverpool HIV checker listed riociguat as a substrate of multiple CYP enzymes and transporters but did not provide interaction severity. Several ART-anti-infective DDIs supported by mechanistic evidence were inconsistently represented across the databases. For example, atazanavir was reported in the literature to inhibit CYP3A4, which may elevate lumefantrine levels and increase the risk of QT prolongation.²⁰ However, this mechanism was not captured in Liverpool HIV, which instead referenced the CYP2B6*6/*6 genotype as a potential modifier. Rifampicin, a potent inducer of CYP3A4 or UGT1A1, was consistently shown to accelerate the metabolism of dolutegravir, efavirenz, darunavir, and ritonavir in the reviewed studies.^21,25,26 Ritonavir, conversely, was found to inhibit P-gp and OATP1B1, suggesting a potential increase in rifampicin plasma concentrations and a heightened risk of hepatotoxicity, an interaction not documented in any of the three databases. The induction of CYP3A4 by efavirenz was reported to affect lumefantrine plasma levels.¹⁹ The nevirapine plasma level decreased from its auto-induction of CYP3A4 when co-administered with rifapentine.²² These mechanisms were generally noted in DrugBank and Drugs.com but were not included in the Liverpool HIV checker.

Table 2.

Possible mechanisms of drug interactions with antiretroviral drugs in a clinical setting: A comparison of drug interaction mechanisms based on evidence from DrugBank, Drugs.com and Liverpool HIV notifications regarding safe combinations.

Author and year	Mechanism	Proposed mechanism from include literature	Noticed in database (✓)			Not noticed in database (x)
Author and year	Mechanism	Proposed mechanism from include literature	DrugBank	Drug.com	Liverpool HIV	DrugBank	Drug.com	Liverpool HIV
Nyangulu et al., 2023¹⁹	Enzyme induction (CYP3A4)	Efavirenz induced CYP3A4-mediated metabolism of lumefantrine, leading to lower plasma levels of the active drug (desbutyl-lumefantrine).	✓	✓	✓
Nyangulu et al., 2023¹⁹	Enzyme induction (CYP3A4)		(CYP3A4 induction)	(CYP3A4 induction)	(CYP3A4 induction)
Nicolò et al., 2022²¹	Enzyme induction (CYP3A4)	Rifampicin induces CYP3A4, leading to a decrease in plasma levels of ritonavir-boosted darunavir and an increased metabolism of ritonavir in PBMCs.	✓	✓	✓	x	x	x
Nicolò et al., 2022²¹	Transporter inhibition (OATP1B1, P-gp)	Ritonavir inhibits P-gp and OATP1B1, increasing rifampicin levels in plasma and potentially causing hepatotoxicity.	(CYP3A4 induction)	(CYP3A4 induction)	(CYP3A4 induction)	(P-gp and OATP1B1 inhibition)	(P-gp and OATP1B1 inhibition)	(P-gp and OATP1B1 inhibition)
Podany et al., 2021²²	Enzyme induction (CYP3A4)	Nevirapine auto-induced CYP3A4, while rifapentine also induced CYP3A4, leading to an increase in nevirapine metabolism.	✓	✓				x
Podany et al., 2021²²	Enzyme inhibition (CYP3A4)		(CYP3A4 induction and inhibition)	(CYP3A4 induction and inhibition)				(CYP3A4 induction and inhibition)
Gausi et al., 2020²³	Genetic variations (NAT2 and CYP2B6 normal, intermediate and slow metabolizers)	Pregnancy induces several physiological changes, such as increased plasma volume, altered body weight and composition, and enhanced hepatic blood flow, which lead to increased clearance of drugs like isoniazid and efavirenz.				x	x	x
	Enzyme inhibition (CYP2B6)	Pregnancy could also alter the activity of drug-metabolizing enzymes. For example, there is an increase in the activity of CYP2B6, which is involved in the metabolism of efavirenz.				(CYP2B6/NAT2 polymorphism during pregnancy and CYP2B6 inhibition)	(CYP2B6/NAT2 polymorphism during pregnancy and CYP2B6 inhibition)	(CYP2B6/NAT2 polymorphism during pregnancy and
		Genetic polymorphisms significantly affect drug metabolism. Normal metabolizers of CYP2B6 and rapid acetylators of NAT2 tend to have lower drug concentrations, increasing their risk of sub-therapeutic levels during pregnancy.
		Isoniazid inhibits the metabolism of efavirenz, particularly in intermediate and slow metabolizers of CYP2B6, thus increasing the risk of efavirenz-related toxicities.
Usman et al., 2022²⁰	Enzyme inhibition (CYP3A4)	Atazanavir inhibits the CYP3A4 isoenzyme, which metabolizes lumefantrine, potentially leading to increased plasma concentrations of lumefantrine and QT prolongation.	✓	✓	✓			x
		The observed QT prolongation may not be solely due to the increased concentration of lumefantrine, but rather to the combined cardiotoxic effects of both drugs, which may act additively.	(CYP3A4 inhibition)	(CYP3A4 inhibition)	CYP2B66/6			(CYP3A4 inhibition)
DeJesus et al., 2019²⁴	Enzyme inhibition (CYP1A1)	Abacavir, dolutegravir, or lamivudine inhibited CYP1A1 and P-gp/BCRP, leading to slower metabolism of riociguat.	✓	✓	✓	x	x	x
	Transporter inhibition (P-gp/BCRP)		(P-gp inhibition, only, dolutegravir)	(CYP1A1 inhibition, only, abacavir)	Only riociguat is metabolized by CYP1A1, CYP3A4, CYP2C8, CYP2J2 and is a substrate of P-gp and BCRP.	(CYP1A1 and P-gp/BCRP inhibition in other drugs)	CYP1A1 and P-gp/BCRP inhibition in other drugs)	(CYP1A1 and P-gp/BCRP inhibition in ART drugs)
Sekaggya-Wiltshire et al., 2023²⁵	Enzyme induction (CYP3A4 and UGT1A1)	High doses of rifampicin induced CYP3A4 and UGT1A1, leading to accelerated metabolism of efavirenz.	✓	✓	✓
			(CYP3A4 and UGT1A1 induction)	(CYP3A4 and UGT1A1 induction)	(CYP3A4 and UGT1A1 induction)
Kengo et al., 2023²⁶	Enzyme induction (intestinal enzymes/transport proteins)	Rifampicin induced intestinal enzymes and/or transport proteins, leading to an increase in pre-systemic elimination of dolutegravir.	✓	✓	✓
			(CYP3A4 and UGT1A1 induction)	(CYP3A4 and UGT1A1 induction)	(CYP3A4 and UGT1A1 induction)

In addition, pharmacogenetic variability and physiological changes during pregnancy were shown to alter ART metabolism significantly. Increased activity of CYP2B6 and NAT2 in normal metabolizers and rapid acetylators was associated with faster clearance of efavirenz and isoniazid, potentially leading to subtherapeutic drug exposure.²³ Conversely, in CYP2B6 intermediate and slow metabolizers, isoniazid inhibited efavirenz metabolism, resulting in elevated plasma levels and an increased risk of neurotoxicity. This genotype-dependent mechanism was not documented in any of the databases reviewed, highlighting the limited integration of pharmacogenomic data in current clinical decision support tools.

Discussion

ART remains essential for improving long-term health outcomes in PWH.^29–34 This study offers a novel contribution to understanding the complexities of DDIs between ART and treatments for comorbidities such as TB, malaria, and pulmonary hypertension. This study identified frequent and severe DDIs by including pharmacokinetic, pharmacodynamic, and pharmacogenomic mechanisms, emphasizing the need for more personalized treatment strategies. We evaluated three online drug-interaction databases (DrugBank, Drugs.com, and the Liverpool HIV database), finding that they inadequately represent key interactions, particularly those influenced by genetic factors.

A key finding is the significant interaction between commonly used ART agents, including dolutegravir and efavirenz, and treatments for comorbid conditions. These drugs are standard components of HIV therapy globally.^1,2 High-dose rifampicin has been shown to reduce plasma concentrations of both dolutegravir and efavirenz.^25,26 However, the effect on dolutegravir was less pronounced,²¹ suggesting that ART regimens containing dolutegravir may require fewer adjustments in HIV/TB co-infected patients. Doubling the dolutegravir dose showed that it could counteract this interaction and maintain adequate drug levels.²⁶ The combination of nevirapine with rifapentine and isoniazid has also been associated with increased viral load and HIV resistance,²² further emphasizing the need for vigilant clinical monitoring to prevent treatment failure. Similarly, efavirenz combined with artemether-lumefantrine significantly lowers lumefantrine and its active metabolite levels, raising concerns about malaria treatment failure and drug resistance in Plasmodium falciparum.¹⁹ Furthermore, coadministration of atazanavir-ritonavir with artemether-lumefantrine has been linked to QT interval prolongation,²⁰ a serious concern in HIV patients with malaria comorbidity. These risks necessitate regular monitoring of liver function, neuropsychiatric status, and electrocardiograms, in cooperation with WHO and U.S. Department of Health and Human Services (HHS) guidelines.^35,36 Additionally, the discovery of interactions between riociguat and ART drugs with dolutegravir, abacavir, or lamivudine in patients with pulmonary hypertension presents further clinical challenges. This combination may lead to stomach cramps, requiring close supervision.²⁴ This study noted how administered drugs alter the pharmacokinetics of ART agents. Rifampicin significantly reduced the AUC for dolutegravir and efavirenz.^25,26 Conversely, combinations such as tenofovir disoproxil fumarate and dolutegravir with islatravir showed no meaningful pharmacokinetic interactions,²⁷ indicating their potential suitability for dual therapy regimens. Islatravir does not inhibit major metabolic enzymes or transporters (e.g., CYP1A2, 2B6, 2C9, UGT1A1), making it a promising candidate for future treatment, though it remains under clinical investigation.^27,37

Pharmacodynamic interactions were also notable. Efavirenz reduced plasma concentrations of lumefantrine, potentially compromising malaria treatment in HIV-malaria co-infection.¹⁹ Similarly, the interaction between atazanavir-ritonavir and artemether-lumefantrine elevated cardiotoxic risks due to QT prolongation.²⁰ These studies showed the critical importance of evaluating both pharmacokinetic and pharmacodynamic factors when managing HIV in patients with concurrent infections.

Our study revealed evidence pointing to the role of pharmacogenomics in modulating ART-related drug interactions. Genetic polymorphisms (i.e., CYP2B6 and NAT2) affect the metabolism of efavirenz and isoniazid, increasing the risk of ADRs.²³ It is particularly crucial in pregnant women, where normal metabolizers of CYP2B6 and rapid acetylators of NAT2 may experience subtherapeutic levels of these drugs. In contrast, CYP2B6 slow metabolizers are at risk of severe ADRs such as neurotoxicity and hepatotoxicity. Genotyping patients before treatment could support safer, more effective ART dosing, especially during pregnancy, when hormonal changes further alter drug metabolism. The effect of rifampicin on dolutegravir bioavailability, despite dose adjustments, was not associated with UGT1A1 polymorphisms but rather with the induction of intestinal enzymes or transport proteins.²⁶ Although no significant pharmacogenetic influence was observed on the rifampicin-efavirenz interaction,²⁹ the findings emphasize the need to incorporate pharmacogenomic testing into HIV treatment, especially in complex clinical scenarios.

This study evaluated drug interaction mechanisms involving ART and compared interaction data across DrugBank, Drugs.com, and the Liverpool HIV interaction database to assess their consistency and completeness. While three databases support clinicians, pharmacists, and researchers in ensuring the safe and effective use of medications, particularly in managing complex ART regimens for PWH, notable discrepancies were identified. For instance, abacavir, lamivudine, dolutegravir, and riociguat inhibited CYP1A1 and P-gp/BCRP, which was not uniformly reported across the databases.²⁴ While abacavir was listed in Drugs.com as a CYP1A1 inhibitor, and dolutegravir was shown in DrugBank to inhibit P-gp, the Liverpool HIV database only identified riociguat as a substrate of CYP1A1, CYP3A4, CYP2C8, CYP2J2, and P-gp/BCRP, without acknowledging interactions with the ART agents. This discrepancy showed a lack of harmonization between mechanistic predictions and clinical documentation. Another key example involves ritonavir, which is known to inhibit P-gp and OATP1B1, thereby increasing plasma concentrations of rifampicin and resulting in a risk of hepatotoxicity.²¹ Surprisingly, this interaction was not documented in any of the three databases. However, all three did acknowledge that rifampicin induces CYP3A4, leading to decreased plasma levels of ritonavir-boosted darunavir and enhanced ritonavir metabolism. Despite the availability of DDIs from databases, these tools do not always account for genetic variability or physiological changes in pregnancy.²³ During pregnancy, increasing hepatic blood flow and plasma volume accelerated the clearance of isoniazid and efavirenz, possibly compromising therapeutic efficacy. Genetic polymorphisms of slow or intermediate metabolizer variants in CYP2B6 can also impair efavirenz metabolism. Coadministration of isoniazid, which inhibits efavirenz metabolism, can further elevate plasma concentrations and increase the risk of toxicity.²³ Prospective studies would be needed to evaluate the real-world impact of ART-related DDIs across diverse populations, considering genetic variability, physiological changes, and comorbidities. This study has several limitations. The review protocol was not registered in PROSPERO. Only 11 clinical studies were included, and they varied in design, sample size, and reporting quality, resulting in some heterogeneity. Some studies also had a high risk of bias in blinding and outcome reporting, which may affect the robustness of certain conclusions. We excluded observational studies, which may have omitted additional relevant evidence. Our analysis was restricted to DrugBank, Drugs.com, and the Liverpool HIV database; other resources such as Micromedex, Lexicomp, and DIDB^® were not included due to limited access. The exclusion of DIDB®, in particular, may have limited our ability to capture more detailed mechanistic insights into transporter- and metabolism-based interactions. Nevertheless, the databases used are widely accessible and provide clinically relevant information for ART-related drug interactions. Most included trials involved small populations, limiting the applicability of results to broader populations. Despite these limitations, our study provided an overview of clinically important DDIs and highlights the need for further research and improved database integration. In addition, although 11 high-quality studies were included, our review did not perform a formal meta-analysis due to differences in outcomes, populations, and pharmacokinetic parameters. As a result, quantitative estimates of ART-related DDIs could not be provided. Future research incorporating individual participant data meta-analysis may strengthen the precision and clinical relevance of these findings.

Conclusion

This study demonstrated critical gaps in current drug interaction databases when managing ART with comorbidities such as TB, malaria, and pulmonary hypertension in PWH. Our findings indicate that DDIs occur across all classes of ART regimens. Non-nucleoside reverse transcriptase inhibitors (efavirenz, nevirapine) frequently interact with tuberculosis drugs (rifampicin, rifapentine, isoniazid) and antimalarials (artemether-lumefantrine), which may reduce efficacy and increase toxicity. Similar to evidence, protease inhibitors (atazanavir-ritonavir, darunavir-ritonavir) interact with rifampicin or artemether-lumefantrine. Integrase inhibitor (dolutegravir) had generally manageable interactions with high-dose rifampicin. In contrast, nucleoside/nucleotide reverse transcriptase inhibitors (tenofovir disoproxil fumarate, islatravir) exhibit minimal interactions and maintain effective HIV suppression. Discrepancies among DrugBank, Drugs.com, and the Liverpool HIV database may hinder optimal clinical decision-making, particularly in settings of polypharmacy. The influence of pharmacogenetic factors, including CYP2B6 and NAT2 polymorphisms, is important for personalized approaches to ART, especially in pregnant individuals. Ongoing updates to interaction databases and integration of pharmacogenomic data are essential to support safer, more effective, and individualized HIV care globally. Future studies should include larger, multicenter trials, broader populations, and expanded database evaluations to optimize real-world outcomes and improve evidence-based ART management.

Supplemental Material

Supplemental Material - Insights from systematic reviews (2019–2024) and drug interaction database analysis in people with HIV and comorbidities

Supplemental Material for Insights from systematic reviews (2019–2024) and drug interaction database analysis in people with HIV and comorbidities by Cherdsak Boonyong, Phattarapon Boonsing, Laphatsanan Ruangritpassorn, Phasapon Thayutichayaton, Phomin Rattanamungkun, Nattharika Chaiboonrueang, Pannapa Powthong in International Journal of STD & AIDS

Supplemental Material

Supplemental Material - Insights from systematic reviews (2019–2024) and drug interaction database analysis in people with HIV and comorbidities

Supplemental Material

Supplemental Material - Insights from systematic reviews (2019–2024) and drug interaction database analysis in people with HIV and comorbidities

Footnotes

ORCID iD

Cherdsak Boonyong

Author contributions

Cherdsak Boonyong was responsible for the study’s concept and design, data collection, data analysis and interpretation, manuscript drafting, critical revision of the manuscript, and final approval. Laphatsanan Ruangritpassorn, Phasapon Thayutichayaton, Phomin Rattanamungkun, Phattarapon Boonsing, and Nattharika Chaiboonrueang contributed to data collection and the analysis of the data. Pannapa Powthong also contributed to the critical revision of the manuscript, review and final approval. All authors are accountable for all aspects of the work, ensuring its integrity and accuracy.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Gandhi

Bedimo

Hoy

, et al. Antiretroviral drugs for treatment and prevention of HIV infection in adults: 2022 recommendations of the international antiviral society-USA panel. JAMA 2023; 329: 63–84.

Trickey

Sabin

Burkholder

, et al. Life expectancy after 2015 of adults with HIV on long-term antiretroviral therapy in Europe and North America: a collaborative analysis of cohort studies. Lancet HIV 2023; 10: 295–307.

Maranchick

Kwara

Peloquin

. Clinical considerations and pharmacokinetic interactions between HIV and tuberculosis therapeutics. Expet Rev Clin Pharmacol 2024; 17: 537–547.

Parra-Rodriguez

Sahrmann

Butler

, et al. Antiretroviral therapy and cardiovascular risk in people with HIV in the united States-An updated analysis. Open Forum Infect Dis 2024; 11: 485.

Obeagu

Ubosi

, et al. Concurrent management of HIV and malaria: a comprehensive review of strategies to enhance quality of life. Medicine (Baltim) 2024; 103: 37649.

Schreck

Istvan

Guerlais

, et al. New psychoactive substances, new behaviours, new drug-drug interactions: pharmacology of a Slam Session. Curr Drug Metabol 2023; 24: 80–91.

Cerrone

Bracchi

Wasserman

, et al. Safety implications of combined antiretroviral and anti-tuberculosis drugs. Expet Opin Drug Saf 2020; 19: 23–41.

van Luin

Van der Ende

Richter

, et al. Lower atovaquone/proguanil concentrations in patients taking efavirenz, lopinavir/ritonavir or atazanavir/ritonavir. AIDS 2010; 24: 1223–1226.

Amariles

Rivera-Cadavid

Ceballos

. Clinical relevance of drug interactions in people living with human immunodeficiency virus on antiretroviral therapy-update 2022: systematic Review. Pharmaceutics 2023; 15: 2488.

10.

Hughes

Waldron

Bennett

, et al. Prevalence of drug-drug interactions in older community-dwelling individuals: a systematic review and meta-analysis. Drugs Aging 2023; 40: 117–134.

11.

Sánchez-Valle

Correia

Camacho-Artacho

, et al. Prevalence and differences in the co-administration of drugs known to interact: an analysis of three distinct and large populations. BMC Med 2024; 22: 166.

12.

Kong

Pozen

Anastos

, et al. Non-HIV comorbid conditions and polypharmacy among people living with HIV age 65 or older compared with HIV-negative individuals age 65 or older in the United States: a retrospective claims-based analysis. AIDS Patient Care STDS 2019; 33: 93–103.

13.

Chen

Tang

, et al. Use of comedications and potential drug-drug interactions in people living with HIV in China. J Infect Chemother 2020; 26: 722–728.

14.

Punekar

Parks

Joshi

, et al. Effectiveness and safety of dolutegravir two-drug regimens in virologically suppressed people living with HIV: a systematic literature review and meta-analysis of real-world evidence. HIV Med 2021; 22: 423–433.

15.

Back

Marzolini

. The challenge of HIV treatment in an era of polypharmacy. J Int AIDS Soc 2020; 23: 25449.

16.

Gandhi

Landovitz

Sax

, et al. Antiretroviral drugs for treatment and prevention of HIV in adults: 2024 recommendations of the international antiviral society-USA panel. JAMA 2025; 333: 609–628.

17.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. PLoS Med 2021; 18: 1003583.

18.

Higgins

JPT

Thomas

Chandler

, et al. Cochrane handbook for systematic reviews of interventions (2nd ed). Wiley-Blackwell, 2019.

19.

Nyangulu

Mungwira

Divala

, et al. Artemether-lumefantrine efficacy among adults on antiretroviral therapy in Malawi. Malar J 2023; 22: 32.

20.

Usman

Oreagba

Busari

, et al. Evaluation of cardiotoxicity and other adverse effects associated with concomitant administration of artemether/lumefantrine and atazanavir/ritonavir-based antiretroviral regimen in patients living with HIV. SPJ 2022; 30: 605–612.

21.

De Nicolò

Calcagno

Motta

, et al. The effect of rifampicin on darunavir, ritonavir, and dolutegravir exposure within peripheral blood mononuclear cells: a dose escalation study. Antimicrob Agents Chemother 2022; 66: 0013622.

22.

Podany

Leon-Cruz

Hakim

, et al. Nevirapine pharmacokinetics in HIV-infected persons receiving rifapentine and isoniazid for TB prevention. J Antimicrob Chemother 2021; 76: 718–721.

23.

Gausi

Wiesner

Norman

, et al. Pharmacokinetics and drug-drug interactions of isoniazid and efavirenz in pregnant women living with HIV in high TB incidence settings: importance of genotyping. Clin Pharmacol Ther 2021; 109: 1034–1044.

24.

DeJesus

Saleh

Cheng

, et al. Pharmacokinetic interaction of riociguat and antiretroviral combination regimens in HIV-1-infected adults. Pulm Circ 2019; 9: 2045894019848644.

25.

Sekaggya-Wiltshire

Nabisere

Musaazi

, et al. Decreased dolutegravir and efavirenz concentrations with preserved virological suppression in patients with tuberculosis and human immunodeficiency virus receiving high-dose rifampicin. Clin Infect Dis 2023; 76: 910–919.

26.

Kengo

Nabisere

Gausi

, et al. Dolutegravir pharmacokinetics in Ugandan patients with TB and HIV receiving standard-versus high-dose rifampicin. Antimicrob Agents Chemother 2023; 67: 0043023.

27.

Rudd

Zhang

Fillgrove

, et al. Lack of a clinically meaningful drug interaction between the HIV-1 antiretroviral agents islatravir, dolutegravir, and tenofovir disoproxil fumarate. Clin Pharmacol Drug Dev 2021; 10: 1432–1441.

28.

Aklillu

Zumla

Habtewold

, et al. Early or deferred initiation of efavirenz during rifampicin-based TB therapy has no significant effect on CYP3A induction in TB-HIV infected patients. Br J Pharmacol 2021; 178: 3294–3308.

29.

Atwine

Baudin

Gelé

, et al. Effect of high-dose rifampicin on efavirenz pharmacokinetics: drug-drug interaction randomized trial. J Antimicrob Chemother 2020; 75: 1250–1258.

30.

Ardsorn

Namjuntra

Kimsungnoen

. Quality of care for people with HIV infection and AIDS receiving antiretroviral therapy in Bamrasnaradura Infectious Diseases Institute. Disease Control Journal 2018; 44: 282–292.

31.

Asavapiriyanont

Kasiwat

. Prevalence of low birthweight infants in HIV-infected women delivered in Rajavithi Hospital. J Med Assoc Thail 2011; 94 Suppl 2: 66–70.

32.

Duangrithi

Polsracoo

Bhuddhataweekul

. Metabolic abnormalities among HIV-infected patients: the rational of national health security for people living with HIV. JPTCP 2020; 27: 78–87.

33.

Siraprapasiri

Tharavichitkul

Suntornpong

, et al. Effects of radiation therapy on immunological and virological status in HIV-infected cancer patients in Thailand: a multicenter prospective study. J Med Assoc Thail 2016; 99: 9–16.

34.

Sriondee

Nateetanasombat

Namjuntra

, et al. Effects of cognitive and behavioral program on drug adherence, CD4 and AIDS knowledge in late adolescent and early adult HIV/AIDS patients. Siriraj Medical Journal 2015; 67: 14–19.

35.

U.S. Department of Health and Human Services (HHS) . Guidelines for the use of antiretroviral agents in adults and adolescents with HIV. HHS, 2024.

36.

World Health Organization (WHO) . Consolidated Guidelines on HIV Treatment, (2nd ed). WHO, 2021.

37.

Gillespie

Reddy

Burris

, et al. Islatravir: evaluation of clinical development for HIV and HBV. Expet Opin Invest Drugs 2024; 33: 85–93.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.06 MB

0.12 MB

0.16 MB