https://orcid.org/0000-0001-9807-0258
Figures
Abstract
Objective
Low-level viraemia (LLV) occurs in some people with HIV (PWH) receiving antiretroviral therapy (ART) and has been linked to inferior treatment outcomes. We investigated factors associated with LLV in a nationwide cohort of Swedish PWH starting ART.
Methods
Participants were identified from the InfCareHIV register, with the following inclusion criteria: ART initiation 2006–2017, age >15 years, ≥4 viral load (VL) results available and no documented treatment interruptions or virologic failure (≥2 consecutive VL ≥200 copies/ml) during follow-up. Starting from 6 months after ART initiation, participants were followed for 24 months and categorised as viral suppression (VS; VL <50 copies/ml) or LLV (≥2 consecutive VL 50–199 copies/ml). We analysed the association between the following factors and LLV using multivariable logistic regression: sex, age, pre-ART VL and CD4 count, ART regimen, country of birth, HIV-1 subtype and transmission category.
Results
Among 3383 participants, 3132 (92.6%) had VS and 251 (7.4%) had LLV. In univariable analyses, factors associated with LLV were male sex, higher age, lower pre-ART CD4 count, higher pre-ART VL and ART regimen. After adjustment, the following factors were associated with LLV (adjusted odds ratio; 95% confidence interval): male sex (1.6; 1.1–2.3), higher pre-ART VL (2.7; 2.2–3.3), pre-ART CD4 count <200 cells/μl (1.6; 1.2–2.2), protease inhibitor (PI)-based regimen (1.5; 1.1–2.1), non-standard ART (2.4; 1.0–5.5) and injecting drug use (2.0; 1.1–3.7).
Citation: Brattgård H, Björkman P, Nowak P, Treutiger CJ, Gisslén M, Elvstam O (2022) Factors associated with low-level viraemia in people with HIV starting antiretroviral therapy: A Swedish observational study. PLoS ONE 17(5): e0268540. https://doi.org/10.1371/journal.pone.0268540
Editor: Jonathan Michael Schapiro, Sheba Medical Center, ISRAEL
Received: February 1, 2022; Accepted: May 3, 2022; Published: May 17, 2022
Copyright: © 2022 Brattgård et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Data cannot be shared publicly. Data are available from the quality register InfCareHIV for researchers who meet the criteria for access to confidential data. We have prepared a dataset underlying the findings in this manuscript and deposited this file within the steering committee of InfCareHIV (contact via Dr. Christina Carlander, Department of Infectious Diseases, Karolinska University Hospital, Huddinge, SE-14186, Stockholm, Sweden, email: christina.carlander@regionstockholm@se).
Funding: This study was supported by a grant (0825-011 8298; to OE) from Department of Research and Development, Region Kronoberg, Växjö. The funders had no role in study design, data colletion and analysis, decision to publish, or preparation of the manuscript.
Competing interests: I have read the journal’s policy and the authors of this manuscript have the following competing interests: PB has received grants from Swedish State, grants from Region Skåne, grants from Gilead Nordic Fellowship during the conduct of the study, and personal fees from Gilead, outside the submitted work. CT has received honoraria as speaker and/or advisor from GlaxoSmithKline/ViiV, outside the submitted work. MG has received research grants from Gilead Sciences and Janssen-Cilag and honoraria as speaker and/or scientific advisor from Amgen, Biogen, Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline/ViiV, Janssen-Cilag, MSD, Novocure, Novo Nordic and Sanofi, outside the submitted work. OE has received grant from Pfizer, outside the submitted work. The manuscript does not include patents, products in development or marketed products, and none of the authors report employment or consultancy as competing interest for this work. This does not alter our adherence to PLOS ONE policies on sharing data and materials.
Introduction
Although most people with HIV (PWH) receiving antiretroviral therapy (ART) achieve persistent viral suppression, some individuals have low but detectable plasma levels of HIV RNA without meeting criteria for virologic failure, a phenomenon commonly referred to as low-level viraemia (LLV) [1]. LLV could arise from two separate mechanisms: either passive release of viral particles from latently infected cells or as a consequence of ongoing replication [2–4]. LLV confers increased risk of virologic failure [5–9] and has also been associated with all-cause mortality in a recent analysis of the nationwide Swedish HIV cohort [10]. In order to interpret these associations, it is important to understand the factors that may influence development of LLV during ART.
Parameters reflecting HIV disease severity at ART initiation (high viral load [VL] and low CD4 count) have been associated with LLV in some studies [11–13], in which associations with sex and HIV acquisition route were also reported [12, 13]. Apart from host and viral characteristics, treatment-related factors might influence the risk of LLV, with higher risk for protease inhibitor (PI)-based regimens compared with those based on non-nucleoside reverse transcriptase inhibitors (NNRTIs) or integrase strand transfer inhibitors (INSTIs) [11, 14, 15]. However, these studies were conducted before introduction of INSTI-based regimens for first-line ART, and the potential association between ART regimen and risk of LLV thus remains uncertain for contemporary treatment practices.
Here, we aimed to investigate viral, host and treatment-related factors with regard to development of LLV in a nationwide cohort of Swedish PWH followed after ART initiation.
Materials and methods
Patient selection
We used data from the nationwide InfCareHIV register, a quality register encompassing >99% of Swedish PWH [16]. This register contains information on demographics, HIV acquisition route, CD4 counts, VL, HIV-1 subtype, antiretroviral drugs prescribed and results from genotypic resistance testing [17]. We included participants who started ART (≥3 non-booster drugs, of which not all were nucleoside/nucleotide reverse transcriptase inhibitors) January 1, 2006, to December 31, 2017, were >15 years old at ART initiation and had ≥ 4 VL measurements during the period 6–30 months after initiation of ART. Participants with documented treatment interruptions or virologic failure (defined as ≥2 consecutive VL measurements ≥200 copies/ml or any VL ≥1,000 copies/ml) during the 24 months observation period were excluded, as well as individuals with HIV-2 infection. The Regional Ethics Committee of Lund, Sweden, approved the study (2017/1023). No specific consent was deemed required for this study; data were pseudo-anonymized.
Definition of viraemia categories
Participants were divided into two categories based on their longitudinal viraemia profile during the observation period: viral suppression (VS) and LLV. VS was defined as VL <50 copies/ml and LLV was defined as ≥2 consecutive VL measurements of 50–199 copies/ml with >4 weeks interval (or one VL in the range 50–199 and one in the range 50–999 copies/ml, thus not meeting the definition of virologic failure). Persons with isolated VL measurements of 50–999 copies/ml preceded and followed by VL<50 copies/ml were included in the VS category.
Statistical analysis
The following variables were considered: sex, age at ART initiation, pre-ART CD4 count (higher/lower than 200 cells/μl) and VL (modelled logarithmically), route of transmission (heterosexual contact, male-to-male sexual contact, injecting drug use and other), country of birth (Nordic countries [Sweden, Denmark, Norway, Finland and Iceland] yes/no), HIV subtype (B/non-B) and type of initial ART regimen (NNRTI-based, PI-based, INSTI-based and non-standard ART [including regimens with more than one anchor drug or fusion/entry inhibitors]). Separately, we also considered CD4 percentage, rather than absolute count, and the CD4:CD8 ratio.
Data were analysed using binary logistic regression. Variables included in the main multivariable model were sex, age, country of birth, transmission category, pre-ART VL, pre-ART CD4 count and ART regimen. The association between HIV subtype and LLV was analysed separately, with adjustment for country of birth and transmission route, in addition to sex and age. For the multivariable analyses, correlation matrices were made with the included independent variables to check for multicollinearity. To investigate if the same factors were associated with LLV in men and women with HIV, respectively, we separately performed an analysis stratified by sex. Likewise, since the management of HIV has changed considerably during the studied years, we performed an analysis stratified by year of treatment initiation (before/after January 1, 2014, the year when dolutegravir became available in Sweden). Missing data were handled using complete case analysis, and we report the number of missing values per each variable. As a sensitivity analysis, age was also modelled using restricted cubic splines. We used IBM SPSS 27 Statistics for Windows, version 10 (IBM Corp., Armonk, NY, United States) and Stata SE/15.1 for Windows (StataCorp LLC, College Station, TX, United States) and report all odds ratios (OR) with 95% confidence intervals (CI). Fig 2 was constructed using R [18] with the ggplot2 package [19].
Results
Participant characteristics
A total of 5821 persons started ART between January 1, 2006, and December 31, 2017. Of these, 2438 (41.9%) were excluded (Fig 1). Of 251 individuals excluded since they met the definition of virologic failure during the observation period, 16 (6.4%) had LLV before reaching failure. Among the remaining 3383 participants, 3132 (92.6%) had VS, and 251 (7.4%) met the study definition of LLV. Approximately similar proportions of LLV were observed across calendar year of ART initiation (Fig 2). The median age at ART initiation was 38 years and 62.6% of the study participants were male. The most common transmission route was heterosexual contact (52.7%), and acquisition through injecting drugs was reported in 141 cases (4.2%). A majority (62.7%) of the participants had pre-ART CD4 count >200 cells/μl, and the median pre-ART VL was 4.8 log10 copies/ml (Table 1). The median number of VL measurements during the study period after starting ART was 5 for VS and 7 for LLV. The median VL during follow-up was <20 copies/ml for VS and 46 copies/ml for LLV.
Abbreviation: ART, antiretroviral therapy.
Abbreviations: ART, antiretroviral therapy; LLV, low-level viraemia.
Factors associated with LLV
The following factors were associated with LLV in univariable analyses: sex, age, pre-ART CD4 count, pre-ART VL and PI-based and non-standard regimen (compared with NNRTI-based). When adjusting for sex, age, pre-ART CD4 count and VL, transmission category, country of birth and ART regimen, a statistically significant association to LLV was found for male sex (adjusted OR [aOR], 1.6; 95% CI, 1.1–2.3), pre-ART CD4 count <200 cells/μl (aOR, 1.6; 95% CI, 1.2–2.2), high pre-ART VL (aOR, 2.7; 95% CI, 2.2–3.3), PI-based regimen (aOR, 1.5; 95% CI, 1.1–2.1), non-standard ART (aOR, 2.4; 95% CI, 1.0–5.5) and injecting drug use (aOR, 2.0; 95% CI, 1.1–3.7) (Table 2). Modelling age with restricted cubic splines did not change these results.
Replacing absolute CD4 counts with CD4 percentages did not substantially change our conclusions regarding the other variables; having higher CD4 percentage was linked to lower odds of LLV (aOR, 0.98; 95% CI, 0.96–0.99). Having a low CD4:CD8 ratio was associated with higher odds of LLV in univariable analysis, but when adjusted for absolute CD4 counts and the other variables in the full model, this was not statistically significant (aOR, 0.9; 95% CI, 0.6–1.4).
In a subanalysis restricted to participants with available HIV subtype, we observed no statistically significant association between HIV subtype B (versus non-B) and the risk of LLV, neither in univariable analysis or after adjustment for sex, age, country of birth and suspected route of transmission (aOR, 0.8; 95% CI, 0.5–1.3).
Furthermore, in a subanalysis stratified by sex, the association between pre-ART CD4 count and LLV was not observed for females (Table 3). The association between higher pre-ART VL and higher odds of developing LLV was statistically significant in both men and women, and in both sexes starting ART before and after 2014. The increased odds of LLV in injecting drug users was restricted to persons initiating ART before 2014 (Table 4).
Discussion
In this study, based on a nationwide cohort of Swedish PWH, LLV during ART was associated with higher pre-ART VL and lower CD4 count, as well as with male sex, injecting drug use and use of PI-based or non-standard ART regimens.
LLV, defined as VL of 50–199 copies/ml, occurred in 7.4% of participants, illustrating that this phenomenon is relatively common in persons receiving modern ART regimens. Although differences in the definition of LLV, cohort composition and follow-up time preclude direct comparisons between studies, the frequency of LLV in our cohort was comparable to contemporary reports from other high-income settings [11–13].
Our findings of associations between higher pre-ART VL and lower pre-ART CD4 (both absolute counts and percentage) and LLV are in agreement with previous studies [11–13, 15]. Both these factors reflect more advanced HIV infection, which suggests that LLV may be related to a larger HIV reservoir [20], and that passive release of virions from latently infected cells could be an important mechanism for LLV in our study population [2, 3]. Apart from these HIV-specific factors, we also observed that LLV was more common in men than in women. This finding is in line with studies conducted in Spain and Austria [12, 13]. In the Austrian study, associations with sex were stratified by transmission route, which showed that females infected through heterosexual contact had lower risk of LLV compared with men who have sex with men [13]. In line with this previous study [13], we observed higher risk of LLV in people with injecting drug use as their transmission category. However, this association was not statistically significant in the subanalysis of people starting ART after 2014. This might reflect changes in the Swedish HIV epidemic and/or better management of HIV in people who inject drugs during the study period. However, the number of people infected with HIV through injecting drug use was small (4.2% of the study population), so the findings of this subanalysis should be interpreted with caution. In our population, the association between sex and LLV was restricted to persons starting ART after 2014, although the proportions of men and women were similar during the whole study period. It is possible that the association between LLV and sex reflects differences in disease severity at HIV diagnosis between men and women, for instance related to late presentation [21].
Previous studies have found higher risk of LLV among people receiving PI-based compared to NNRTI-based ART [11, 12, 14, 15, 22]. We also observed such an association, but in the subanalyses restricted to participants starting ART before or after 2014, this relationship was not statistically significant. The association between different regimens and viraemia profile during ART might be influenced by indication bias, with higher likelihood of PI prescription in persons with anticipated lower adherence [11, 13]. Yet, the association between PI and LLV has also been observed in a retrospective analysis of two randomized trials [23] and is also supported by increased occurrence of viral blips in PWH on PI-based compared to other regimens [24]. We also observed higher odds of LLV among people with non-standard ART regimens (including combinations of more than one core agent and fusion or entry inhibitors). The number of patients with these regimens was small, and we consider it likely that they reflect a special category. For example, detection of pre-ART drug resistance mutations may have led to use of non-standard initial regimen. Some of these persons could also have undocumented previous ART experience. Bernal et al. reported lower frequency of LLV in persons receiving INSTI-based regimens [12], but in our material, we found no indication that INSTI-based ART confers reduced risk of LLV compared with other regimens.
Since 2015, ART is recommended for all PWH, irrespective of disease severity [25, 26]. In Swedish guidelines, ART has been recommended irrespective of CD4 counts since 2014 [27]. Since high pre-ART VL and low CD4 count were major risk factors for LLV in our material, earlier ART initiation could be expected to result in reduced incidence of LLV. Still, advanced disease at HIV diagnosis remains common in Europe [28], so improved testing strategies are likely needed to consolidate the effect of early ART on the occurrence of LLV on a population level.
The main limitation of this study concerns residual confounding. We adjusted for several potential confounders, but the database did not contain information on some factors which have been associated with LLV in other studies, such as socioeconomic status or treatment adherence [29]. Furthermore, misclassification of viraemia category might have occurred due to relatively infrequent sampling (reflecting clinical practice). For example, it is possible that some individuals categorized as virally suppressed may have had unrecognized shorter periods of LLV. Also, some individuals with LLV were excluded since they developed failure; our results are therefore restricted to people with LLV without treatment failure during the first years after ART initiation. Lastly, while previous data indicate that type of VL assay could be associated with LLV [13], we lack information on which assays that were used for specific VL measurements. The major strength of our study, which is the first to describe factors associated with LLV in Northern Europe, is that it is based on a nationwide cohort of PWH. Furthermore, we present characteristics of LLV among patients starting ART during a period with access to modern INSTI-based regimens.
Conclusion
Among Swedish PWH initiating ART between 2006 and 2017, markers of HIV disease severity at ART initiation were associated with higher risk of LLV. This risk was also increased in men, people who acquired HIV through injecting drug use and in patients receiving PI-based or non-standard ART regimens. Our findings imply that early initiation of ART, which is now recommended for all PWH, could result in reduced occurrence of LLV.
Acknowledgments
This study benefitted from data provided by the national quality register InfCareHIV.
References
- 1. Crespo-Bermejo C, de Arellano ER, Lara-Aguilar V, Valle-Millares D, Gomez-Lus ML, Madrid R, et al. Persistent low-Level viremia in persons living with HIV undertreatment: An unresolved status. Virulence. 2021;12(1):2919–31. pmid:34874239
- 2. Tobin NH, Learn GH, Holte SE, Wang Y, Melvin AJ, McKernan JL, et al. Evidence that low-level viremias during effective highly active antiretroviral therapy result from two processes: expression of archival virus and replication of virus. J Virol. 2005;79(15):9625–34. pmid:16014925
- 3. Van Zyl GU, Katusiime MG, Wiegand A, McManus WR, Bale MJ, Halvas EK, et al. No evidence of HIV replication in children on antiretroviral therapy. J Clin Invest. 2017;127(10):3827–34. pmid:28891813
- 4. Cohen Stuart JW, Wensing AM, Kovacs C, Righart M, de Jong D, Kaye S, et al. Transient relapses ("blips") of plasma HIV RNA levels during HAART are associated with drug resistance. J Acquir Immune Defic Syndr. 2001;28(2):105–13. pmid:11588503
- 5. Elvstam O, Medstrand P, Yilmaz A, Isberg PE, Gisslen M, Bjorkman P. Virological failure and all-cause mortality in HIV-positive adults with low-level viremia during antiretroviral treatment. PLoS ONE. 2017;12(7). pmid:28683128
- 6. Joya C, Won SH, Schofield C, Lalani T, Maves RC, Kronmann K, et al. Persistent Low-level Viremia While on Antiretroviral Therapy Is an Independent Risk Factor for Virologic Failure. Clin Infect Dis. 2019;69(12):2145–52. pmid:30785191
- 7. Fleming J, Mathews WC, Rutstein RM, Aberg J, Somboonwit C, Cheever LW, et al. Low-level viremia and virologic failure in persons with HIV infection treated with antiretroviral therapy. AIDS. 2019;33(13):2005–12. pmid:31306175
- 8. Vandenhende MA, Perrier A, Bonnet F, Lazaro E, Cazanave C, Reigadas S, et al. Risk of virological failure in HIV-1-infected patients experiencing low-level viraemia under active antiretroviral therapy (ANRS C03 cohort study). Antivir Ther. 2015;20(6):655–60. pmid:25735799
- 9. Antiretroviral Therapy Cohort Collaboration, Vandenhende MA, Ingle S, May M, Chene G, Zangerle R, et al. Impact of low-level viremia on clinical and virological outcomes in treated HIV-1-infected patients. AIDS. 2015;29(3):373–83. pmid:25686685
- 10. Elvstam O, Marrone G, Medstrand P, Treutiger CJ, Sonnerborg A, Gisslen M, et al. All-Cause Mortality and Serious Non-AIDS Events in Adults With Low-level Human Immunodeficiency Virus Viremia During Combination Antiretroviral Therapy: Results From a Swedish Nationwide Observational Study. Clin Infect Dis. 2021;72(12):2079–86. pmid:32271361
- 11. Wirden M, Todesco E, Valantin MA, Lambert-Niclot S, Simon A, Calin R, et al. Low-level HIV-1 viraemia in patients on HAART: risk factors and management in clinical practice. J Antimicrob Chemother. 2015;70(8):2347–53. pmid:25921516
- 12. Bernal E, Gomez JM, Jarrin I, Cano A, Munoz A, Alcaraz A, et al. Low-Level Viremia Is Associated With Clinical Progression in HIV-Infected Patients Receiving Antiretroviral Treatment. J Acquir Immune Defic Syndr. 2018;78(3):329–37. pmid:29543636
- 13. Leierer G, Grabmeier-Pfistershammer K, Steuer A, Geit M, Sarcletti M, Haas B, et al. Factors associated with low-level viraemia and virological failure: Results from the Austrian HIV cohort study. PLoS ONE. 2015;10(11). pmid:26566025
- 14. Doyle T, Smith C, Vitiello P, Cambiano V, Johnson M, Owen A, et al. Plasma HIV-1 RNA detection below 50 copies/ml and risk of virologic rebound in patients receiving highly active antiretroviral therapy. Clin Infect Dis. 2012;54(5):724–32. pmid:22238167
- 15. Vancoillie L, Demecheleer E, Callens S, Vogelaers D, Vandekerckhove L, Mortier V, et al. Markers associated with persisting low-level viraemia under antiretroviral therapy in HIV-1 infection. J Antimicrob Chemother. 2014;69(4):1098–103. pmid:24335486
- 16. Gisslén M, Svedhem V, Lindborg L, Flamholc L, Norrgren H, Wendahl S, et al. Sweden, the first country to achieve the Joint United Nations Programme on HIV/AIDS (UNAIDS)/World Health Organization (WHO) 90-90-90 continuum of HIV care targets. HIV medicine. 2017;18(4):305–7. pmid:27535540
- 17.
InfCareHIV. Årsrapport InfCareHIV 2020 [cited 2022 April 1]. https://qrcstockholm.se/wp-content/uploads/2021/10/%C3%85rsrapport_InfCareHIV2020.pdf.
- 18.
R Core Team. R: A Language and Environment for Statistical Computing. 2019. https://www.R-project.org/.
- 19.
Wickham H. Ggplot2: elegant graphics for data analysis. New York: Springer; 2009.
- 20. Bachmann N, von Siebenthal C, Vongrad V, Turk T, Neumann K, Beerenwinkel N, et al. Determinants of HIV-1 reservoir size and long-term dynamics during suppressive ART. Nat Commun. 2019;10(1):3193. pmid:31324762
- 21. Op de Coul EL, van Sighem A, Brinkman K, van Benthem BH, van der Ende ME, Geerlings S, et al. Factors associated with presenting late or with advanced HIV disease in the Netherlands, 1996–2014: results from a national observational cohort. BMJ Open. 2016;6(1):e009688. pmid:26729389
- 22. Wiesmann F, Braun P, Knickmann M, Knechten H. Low level HIV viremia is more frequent under protease-inhibitor containing firstline therapy than under NNRTI-regimens. J Int AIDS Soc. 2014;17(4 Suppl 3):19828. pmid:25397572
- 23. Taiwo B, Gallien S, Aga E, Ribaudo H, Haubrich R, Kuritzkes DR, et al. Antiretroviral drug resistance in HIV-1-infected patients experiencing persistent low-level viremia during first-line therapy. J Infect Dis. 2011;204(4):515–20. pmid:21791652
- 24. Sorstedt E, Nilsson S, Blaxhult A, Gisslen M, Flamholc L, Sonnerborg A, et al. Viral blips during suppressive antiretroviral treatment are associated with high baseline HIV-1 RNA levels. BMC Infect Dis. 2016;16:305. pmid:27329293
- 25. Insight Start Study Group, Lundgren JD, Babiker AG, Gordin F, Emery S, Grund B, et al. Initiation of Antiretroviral Therapy in Early Asymptomatic HIV Infection. N Engl J Med. 2015;373(9):795–807. pmid:26192873
- 26. Temprano Anrs Study Group, Danel C, Moh R, Gabillard D, Badje A, Le Carrou J, et al. A Trial of Early Antiretrovirals and Isoniazid Preventive Therapy in Africa. N Engl J Med. 2015;373(9):808–22. pmid:26193126
- 27.
The Swedish Reference Group for Antiviral Therapy (RAV). Antiretroviral behandling av HIV-infektion 2014, uppdaterad version—Behandlingsrekommendation. 2014.
- 28. Late Presentation Working Groups in EuroSida and Cohere. Estimating the burden of HIV late presentation and its attributable morbidity and mortality across Europe 2010–2016. BMC Infect Dis. 2020;20(1):728. pmid:33028235
- 29. Goupil de Bouille J, Collignon M, Capsec J, Guillon L, Le Moal G, Barin F, et al. Low-level HIV viremia is associated with low antiretroviral prescription refill rates and social deprivation. AIDS Care. 2021;33(11):1445–50. pmid:32794406