Determining the acceptability of point-of-care urine tenofovir testing and its performance in predicting HIV RNA suppression

Abstract

Background

The detection of tenofovir (TFV) metabolites by point-of-care (POC) urine lateral flow immunoassays (LFIA) indicates adherence to tenofovir-containing HIV pre-exposure prophylaxis. However, the association between urine TFV metabolites as detected by LFIA and HIV viral load suppression in people receiving TFV-based antiretroviral therapy (ART) is unknown as is patient and clinician acceptability of POC urine LFIA testing in clinical practice in low- and middle-income country settings.

Methods

We enrolled 409 people living with HIV from two HIV clinics in Lesotho and investigated the performance of POC urine LFIA TFV testing in predicting viral suppression. We interviewed 12 study participants and conducted a focus-group discussion with 5 clinicians to gather opinions on POC urine TFV testing.

Results

Using a viral load threshold of 1000 copies/mL, 398 (98%) participants were virologically suppressed, and 8 were viremic. Tenofovir was detected in the urine of 405 (99%) participants. The sensitivity of the POC urine LFIA test in detecting TFV in participants with viral suppression was 99.3% (95% CI: 97.8–99.8); the specificity was 12.5% (95% CI: 0.3–52.6). The positive and negative predictive values were 98.3% and 25%, respectively. Point-of-care urine TFV testing was viewed favorably by both participants and clinicians. However, clinicians stated that the 2–3-days detection window of the assay studied limits adherence categorization.

Conclusions

In our study cohort, urine POC TFV testing demonstrated high sensitivity in predicting viral suppression, but low specificity and negative predictive value. Urine POC TFV testing was highly acceptable to participants and clinicians; however, clinicians expressed concern about its clinical utility because of its limitations. While further research is needed to assess performance in less adherent populations, this test may support adherence counseling in some clinical settings.

Keywords

HIV treatment equipment antiretroviral therapy Africa

Background

As of December 2021, 37.6 million people were living with HIV globally, with 27.4 million receiving antiretroviral therapy (ART).¹ When taken as prescribed, ART is highly effective in blocking HIV replication, halting progression to AIDS and preventing viral transmission. However, suboptimal adherence limits ART effectiveness, leading to treatment failure, onward transmission of virus and potential selection of drug resistance. The early, rapid, and accurate identification by health care providers of patients with suboptimal adherence to ART is necessary to optimize ART treatment outcomes.

Current methods for estimating ART adherence are limited.² Patient self-reports may be inaccurate due to recall or social desirability biases, and pill-counts may be prone to inaccuracies due to manipulation or misclassification resulting from differing medication allocation practices.^2,3 Objective measures of adherence such as antiretroviral drug levels measured in dried blood spots, plasma or hair may overcome these limitations. Use of plasma drug-level measurement, for instance, has been shown to accurately identify non-adherence to HIV treatment, thus guiding targeted efforts to support improved adherence.⁴ However, drug levels measured in blood and hair are currently impractical for routine clinical use due to their high costs and lengthy wait times for results.

Tenofovir (TFV)-disoproxil-fumarate (TDF) is a component of first-line ART regimens recommended by the World Health Organization (WHO). In 2021, WHO recommended the use of dolutegravir (DTG)-based ART as first-line treatment for all adults due to its tolerability, high efficacy, and high barrier to resistance.⁵ TDF is combined with DTG and lamivudine (3TC) in a fixed-dosed tablet formulation (TLD) and provides an effective regimen with minimal pill-burden and a favorable side-effect profile. As of November 2021, it was estimated that 82% of all ART prescribed in countries receiving U.S President’s Emergency Plan for AIDS Relief (PEPFAR) funding was TLD.⁶

Urine can be used as a specimen for drug level testing, with the advantage of being easily and non-invasively obtained. Assays utilizing high-performance liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) have been developed to measure TFV levels in urine. Urine levels of TFV decay in a log-linear fashion, correlating with time since the last dose of the drug. Urine-based LC-MS/MS assay showed superior sensitivity compared to plasma-based assays at detecting TFV within 7 days of the last dose.^7,8 Furthermore, LC-MS/MS has been demonstrated to detect urine TFV levels at ≥1000 ng/mL when concurrent plasma levels were detectable, at up to 3-days after a TFV dose,⁸ establishing this as the urine cut-off for detecting a recent TFV dose within 2–3 days. Unfortunately, LC-MS/MS is expensive and requires specialized equipment and personnel, limiting its real-world applicability, especially in resource-limited settings.

UrSure, Inc has developed an investigational point-of-care (POC) lateral flow immunoassay (LFIA) capable of detecting TFV-specific antibody levels above 650 ng/mL in urine. The 650 ng/mL threshold was established based on internal pharmacokinetic studies to have 100% sensitivity and specificity in detecting adherence within 2–3 days when compared to LC-MS/MS.⁹ The major advantages of this test are its ease of use, rapid results, and its potential low cost.

Measurement of urine TFV levels has been shown to correlate with ART adherence.¹⁰ Urine TFV LFIA could potentially change viral load monitoring practices in low resource settings if validated to show a strong association with viral suppression. As POC urine TFV testing is easier, faster, and cheaper to obtain than plasma viral loads, it could potentially be used as a surrogate, available for more frequent use. A POC LFIA detecting TFV could support early identification of non-adherence, thus enabling direction of resources towards targeted adherence support, thereby reducing the potential for emergence of drug-resistant HIV. A negative urine TFV test may signal need for enhanced adherence counseling prior to viral load testing.

Furthermore, results could be used to determine if an individual with viral non-suppression (defined as ≥1000 copies/mL) should have a drug resistance test. For instance, a positive urine TFV test in the setting of virologic non-suppression may signal reflex HIV drug resistance testing in specific scenarios, if drug resistance genotyping is available. In resource-limited settings, viral non-suppression may lead clinicians to prematurely transition patients to second-line regimens, which may be more costly as well as more challenging to adhere to due to side effects. Urine TFV testing may help inform on adherence patterns and decrease instances of unnecessary regimen switches.

To our knowledge, this is the first reported study of the association of TFV detected by POC LFIA testing and viral suppression in people receiving TFV-containing ART for the treatment of HIV in a low- and middle-income country.

Setting

This study was conducted in the Kingdom of Lesotho at the Sankatana ART clinic and the Butha-Buthe Government Hospital clinic.

The Kingdom of Lesotho has a population of 2.2 million people, with 280,000 people estimated to be living with HIV. As of 2020, an estimated 82% of those living with HIV were receiving ART, with approximately 230,000 achieving suppressed viral loads.¹¹ In Lesotho, ART is provided at 226 clinics following a public health model of care based on national ART guidelines informed by WHO recommendations.^5,12

In July 2019, Lesotho updated its national guidelines to reflect the WHO-recommended transition to TLD as recommended first-line treatment for all adults.¹³ Alternative first-line therapy consists of a combination of TDF and 3TC with efavirenz (EFV). Second-line ART consists of protease-inhibitor-(PI) based therapy. In keeping with WHO guidelines, all adult patients receiving ART have at least one HIV viral load test performed annually.

Adherence to ART is monitored at routine clinic visits utilizing pill counts. Patients identified as having suboptimal adherence are enrolled into an enhanced adherence counselling program, where they have more frequent visits to address identified barriers to adherence.

Methods

Participant enrollment

Sample size determination

To determine sensitivity and specificity in detecting viral suppression, the sample size was determined utilizing a one-sample test of sensitivity and specificity. Based on country data, a conservative estimate of 5% viral non-suppression was expected prior to initiating the study. Utilizing a two-sided significance level of 5%, a sample size of at least 400 participants was estimated to result in at least 98% power in detecting a sensitivity and specificity of 95%, that is statistically different from 50% (no diagnostic ability), utilizing a two-sided binomial test. For the in-depth interviews, we estimated that 10-12 participants would be enough to reach code saturation.

Urine POC testing

All patients attending the Sankatana clinic, as well as patients enrolled in enhanced adherence counselling at the Butha-Buthe clinic, on or after the study start date were screened for eligibility. Participant enrollment occurred between June and October 2021. HIV-infected adults (≥18 years of age) receiving any TFV-containing first- or second-line ART regimen were eligible for the study. Individuals who declined to provide informed consent or who were unable to provide a urine specimen were excluded. Participants who were not enrolled in enhanced adherence counselling had blood drawn for HIV viral load testing, per routine practice. Outside the protocol of this study, participants who were enrolled in enhanced adherence counselling would not usually undergo routine HIV viral load testing, and thus were only enrolled if they also consented to HIV viral load testing upon enrollment. Urine was provided by patients and collected by clinic nurses. Nurses performed the urine POC TFV testing in a private location within the clinic, and testing was performed following manufacturer instructions and using pre-packaged testing equipment. Training on how to perform the urine POC test was provided by study investigators and clinic leadership prior to initiation of the study and following instructional materials provided by UrSure, Inc.

Results from the urine POC test were recorded for study purposes but were not provided to participants or clinicians.

Viral load testing was performed at the Lesotho National Reference and Butha-Buthe Government Hospital Laboratories using HIV RNA real-time polymerase chain reaction per manufacturer instructions (Cobas® 4800 system, Roche).

Baseline demographic data were abstracted from participants medical records including most recent CD4 cell count, previous viral load test result, age, comorbidities, and concurrent medications.

Determining Acceptability of Urine Tenofovir Testing using Qualitative Methods

Semi-structured interview guides were developed and used for individual in-depth interviews with participants and a focus group discussion (FGD) with ART providers. We interviewed a convenience sample of 12 study participants who had completed urine POC TFV testing to obtain information about their attitudes and perceptions about use of POC urine TFV testing during a routine clinic visit. Study participants were asked questions related to their understanding of the urine POC test, acceptability of using it for routine adherence testing, their preferences regarding objective versus subjective adherence measures, and concerns about how an objective test would affect medication adherence discussions with ART providers. All in-depth interviews with study participants were conducted in Sesotho by one study coordinator and participant responses were recorded in Sesotho and subsequently translated into English for analysis.

To establish the values and preferences of ART clinicians regarding urine POC testing, we conducted a FGD with 5 nurse clinicians at the Sankatana ART clinic. Clinicians were asked questions about their perceptions of urine POC testing in the clinic, as well as potential barriers to its implementation. Clinicians were also queried on their understanding of how to interpret the test results and potential significance of the results as they pertain to clinical care. The clinician FGD was conducted entirely in English, audio-recorded, and transcribed for analysis.

Data analysis

The diagnostic performance of the urine POC test for predicting viral suppression was determined by calculating sensitivity, specificity, and positive and negative predictive values for the test. Confidence intervals for sensitivity and specificity were calculated utilizing the Clopper-Pearson interval. Confidence intervals for positive and negative predictive values were calculated utilizing standard logit confidence intervals. Estimates were calculated utilizing CDC Epi Info™ and Microsoft® Excel.

Information obtained from participant interviews and provider FGDs were analyzed separately using English transcripts. Prior to analysis, overarching domains were created to guide the analyses. An initial coding framework was developed by the study team. Transcripts were independently coded by two research team members and organized by themes. The codes and themes were then discussed by two of the investigators. When different codes emerged, these were discussed and added to the coding framework when agreed upon. The transcripts were then re-read with the new codes in mind. The final codes were organized into categories and discussed with the study team to determine the key themes.

Ethics

This study was approved by the Lesotho Ministry of Health Research and Ethics Committee (Reference ID: 129–2020) and the Tufts Health Sciences Institutional Review Board (STUDY00001744).

Results

Urine Point of Care Tenofovir Measurement and HIV Viral Load Testing

A total of 409 participants were enrolled into the study. The median age of participants was 44 years (range: 18–84), with a median previous viral load at time of enrollment of 344 copies/mL (Range: <20–138,000) and a median CD4 cell count of 483 cells/uL (range: 2–1781) (Table 1). The median number of years on ART at enrollment was 7.5 years (range: 0.3–17.5). Most participants were receiving first-line ART regimens (399, 97.6%) (Table 2). There were 22 participants who were enrolled into enhanced adherence counselling.

Table 1.

Participant characteristics at study enrollment, n = 409.

Characteristic	Median (range) or number (%)
Median age (years)	44 (18–84)
Gender
Male	152 (37.16%)
Female	257 (62.84%)
Median previous viral load test result in past year (copies/mL)	344 (<20–138,000)
Median prior CD4 cell count (cells/uL)	483 (2–1781)
Total participants with viral suppression	398 (98.03%)

Table 2.

ART regimens at enrollment.

ART regimen	Number (%)
TDF-3TC-DTG	387 (94.62%)
TDF-3TC-EFV	12 (2.93%)
TDF-AZT-3TC-DTG	2 (0.49%)
TDF-AZT-3TC-LPV/r	1 (0.24%)
TDF-3TC-LPV/r	7 (1.17%)

Abbreviations: TDF, Tenofovir-disoproxil-fumarate; 3TC, Lamivudine; DTG, Dolutegravir; EFV, Efavirenz; AZT, Zidovudine; LPV/r, Ritonavir-boosted Lopinavir; ATV/r, Ritonavir-boosted Atazanavir; DRV/r, Ritonavir-boosted darunavir.

A viral load threshold of <1000 copies/mL was utilized to define viral suppression based on WHO recommendations for low- and middle-income countries.⁵ Viral suppression was achieved by 98% (398/406) of participants at time of enrollment, and 8 were viremic (Table 3). Viral load test results were unavailable for three participants. Of the 8 viremic participants, 7 were receiving TDF-3TC-EFV, and 6 of 7 had detectable TFV in the urine. In total, TFV was detected in the urine of 99% (405/409) of study participants, with only 4 having no TDF detected in the urine. Of the 3 participants with both negative urine POC test result and virologic suppression, 2 were receiving TLD. The sensitivity of detecting TFV by urine POC testing in participants with virological suppression was 99.3% (95% CI: 97.8–99.8), while the specificity was 12.5% (95% CI: 0.3–52.6). The positive predictive value was 98.3% and the negative predictive value was 25%.

Table 3.

Urine point of care tenofovir measurement and HIV viral RNA results.

		HIV viral RNA
		Suppressed (<1000 copies/mL)	Non-suppressed (≥1000 copies/mL)	Total
Urine point of care tenofovir measurement	Tenofovir detected	395	7	402
Urine point of care tenofovir measurement	Tenofovir not detected	3	1	4
	Total	398	8	406

Using a stricter definition of viral suppression (≤200 copies/mL), 97.3% of participants had viral suppression, yielding a sensitivity, specificity, positive and negative values of 99.2%, 9%, 97.5% and 25%, respectively.

Study participant interviews

When asked how well they take their ART, nearly all the participants reported that they were generally adherent but occasionally missed doses due to work-related issues. When asked how comfortable they felt telling their doctors about their missed doses, only one participant admitted feeling uncomfortable telling the truth because “sometimes health providers may be harsh when we tell them the truth about missed doses, so we end up lying”. The remaining participants expressed that they felt it was important for their doctors to know the truth about their adherence or non-adherence to effectively direct their care, with one representative participant stating, “hiding information will hinder their potential to help me”.

All participants reported favorable attitudes towards the test, expressing that the test was important and more reliable than self-reported adherence or pill counts. All stated that they would believe the results of the urine POC test as they would other clinical test results. Many also mentioned the rapid nature of the test as an advantage.

Most participants reported that they would change their behavior if such testing were to be implemented into routine clinical practice, stating they would either be reminded to start taking their medications prior to a visit or feel more motivated to remain adherent between clinical visits. Two participants admitted to occasionally discarding pills to manipulate pill counts and thought that such a test could change their behavior, with one participant stating, “It will definitely (change my behavior) because I will likely take my medication correctly as I know that even if I can lie through pill tossing, the test will prove me otherwise”.

All participants interviewed expressed enthusiasm should urine POC testing be integrated into routine clinic visits. No participant reported any concerns or hindrances about providing urine and expressed willingness to take this test if recommended by their provider in the future.

Clinician focus group discussion

During FGDs, clinicians opined that the test was easy to perform and not time consuming. Clinicians reported that any clinical staff, such as a doctor or nurse, could perform the test. They expressed that this test would be best performed during routine clinical visits, rather than at visits to pick up medications, as participants would be more receptive to allocating time to perform the test.

Some clinicians felt that the test was not intuitive to interpret. They also expressed a desire for a semi-quantitative test that showed varying color intensity depending on levels of TFV detected in the urine. Clinicians also stated that participants may feel unhappy when learning of a test result that was negative for TFV.

The main concern clinicians had regarding the utility of the test was that the test only reflected adherence over the preceding 2–3 days. Clinicians expressed that should this test be implemented as part of routine practice, patients could start taking medication shortly before a clinic visit despite not being generally adherent, thus limiting the clinical utility of the results.

Discussion

In the studied population, results showed a high sensitivity and positive predictive value of a positive POC urine TFV LFIA in predicting viral suppression; however, poor specificity and negative predictive value was observed. Our sample size calculation assumed at least 5% viral non-suppression; however, due to the extremely high levels of viral suppression of patients at the participating clinics, and Lesotho at large,¹¹ only a limited number of participants with viral non-suppression were able to be recruited into this study. A recent study assessing viral suppression in Lesotho after the initiative to transition to the use of TLD, revealed 98% suppression after regimen switch.¹⁴ Thus, the generalizability of results from this study is limited to populations with similar prevalence of viral non-suppression.

In this setting, while a positive urine POC TFV test may be suggestive of viral suppression in these populations, its low negative predictive value negates the test’s ability to completely replace viral load testing, even when TFV is detected. Future studies of POC urine TFV testing are warranted including assessment of test’s performance characteristics when populations with known prior viral suppression are tested repeatedly and at regular intervals– results may demonstrate that repeated detection of TDF could allow for future deferral of viral load testing.

Acceptability of the test among study participants was highly favorable, with most viewing the POC urine TFV LFIA as preferable and superior to current adherence monitoring measures based on self-reports and pill counts. Moreover, participants reported a high degree of confidence in test results and the theoretical benefits of its use in promoting adherence. Clinicians and HIV-positive study participants also favored that the urine POC test was quick and easy to perform in the clinic.

Participants stated that knowing that their urine could be checked for TFV would change their behavior and increase adherence. While this may result in a positive change, it also presents a potential significant limitation to adherence monitoring by such methods, known as “white coat adherence”.¹⁵ With “white coat adherence,” patients may temporarily improve adherence in the days prior to anticipated testing or clinic visits, resulting in them being incorrectly classified as adherent.

Clinicians expressed that the most appropriate time to perform this test was during scheduled clinical visits, as patients would be more willing to invest time toward it, rather than, during visits to the on-site pharmacy for pick-ups of drug refills. However, this narrower use of testing would result in fewer opportunities to randomly test patients, potentially resulting in the test being more predictable and subject to forms of manipulation.

Amongst clinicians who participated in the focus group discussion, the 2–3-days window was seen as the major limiting factor against implementation of this test in the clinical setting, with one participant summarizing the problem as “the results are the same for someone who took the medication for 30 days and someone who took the medication for the last 3 days”. To be better able to distinguish between these patterns of adherence, focus group participants expressed the desire for a semi-quantitative method which would distinguish the length of time of consecutive adherence. A urine TFV LFIA which produces semi-quantitative results, where color intensity delineates urine TFV levels was recently validated in people living without HIV and taking TFV-containing ART.^16,17 However, like other currently available urine TFV LFIAs, TFV is detected only within 4 days of last dose. As a result, the short detection window similarly limits this assay’s clinical utility. The validation of an assay sensitive enough to semi-quantitatively assess adherence over a longer window may have higher specificity and clinical relevance in populations receiving ART. Further studies are needed to determine the appropriate window of detection for urine TFV that would also correlate with viral load suppression.

The qualitative nature of the UrSure test, and the brief window of TFV detection of within 2–3 days greatly limits its ability to differentiate between intermittent or sustained adherence. However, in populations receiving TFV-containing ART, the absence of TFV in the urine using a test with a short window of detection may signal suboptimal adherence, providing the opportunity for targeted adherence support.

Among participants receiving enhanced adherence counselling, only 8 of 22 had viral non-suppression. Outside the protocol of this study, patients receiving enhanced adherence counseling in Lesotho only receive viral load testing after having achieved a period of sustained adherence, of at least 3 months, as determined by pill counts. However, most participants receiving enhanced adherence counselling in our study were found to have virologic suppression and a urine POC test positive for TFV, highlighting another potential use of this test, which is to support adherence counseling efforts.

Amongst participants, 95% were receiving a dolutegravir-containing regimen, limiting the ability to assess performance of the urine TFV LFIA with second-line regimens or with alternative EFV-containing first-line regimens. Alternative EFV-based first-line regimens have a lower genetic barrier to selection of drug resistance, and the presence of drug-resistant HIV has been shown to be a cause for discrepant results in patients taking EFV-based ART with viremia and a positive urine TFV test.¹⁸ DTG-containing regimens are generally more forgiving of imperfect adherence allowing more leeway for rapid viral re-suppression without emerging resistance.¹⁹ It is notable that the majority of viremic participants in our study were taking EFV-based regimens despite having detectable TFV in the urine. In contrast, 2 of 3 participants with no detected TFV in the urine and virologic suppression were receiving DTG-containing regimens.

In conclusion, this study supports high clinician and patient acceptability of urine POC TFV testing, as well as demonstrates high positive predictive value in detecting viral load suppression. However, in populations with high levels of virologic suppression, the urine TFV LFIA assessed in this study demonstrates poor specificity and negative predictive value. This test has value in supporting adherence counseling, though demonstrates significant limitations with respect to specificity that prevent it from replacing viral load testing; however, future efforts aimed at developing a urine LFIA with semi-quantitative properties and a longer window of discrimination may enhance the ability to predict viral non-suppression.

Footnotes

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.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Stuart B. Levy Center for Integrated Management of Antimicrobial Resistance and the Sherwood L. Gorbach, M.D., M62, Endowed Research Fund. The urine tenofovir lateral flow immunoassay devices were donated by UrSure, Inc. UrSure did not participate in the study design, data collection, analysis, interpretation of findings, or in the production of this manuscript. UrSure provided instructional directions for equipment use and were present for equipment training.

ORCID iD

Terry A Marryshow

References

UNAIDS 2021 . Epidemiological Estimates [Internet]. Unaids.org, 2022 [cited 2022 Jan 16]; Available from: https://www.unaids.org/sites/default/files/media_asset/UNAIDS_FactSheet_en.pdf

Spinelli

Haberer

Chai

, et al. Approaches to objectively measure antiretroviral medication adherence and drive adherence interventions. Curr Hiv/aids Rep 2020; 17(4): 301–314.

Biressaw

Abegaz

Abebe

, et al. Adherence to antiretroviral therapy and associated factors among HIV infected children in Ethiopia: unannounced home-based pill count versus caregivers’ report. BMC Pediatr 2013; 13(1): 132.

Landovitz

Beymer

Kofron

, et al. Plasma tenofovir levels to support adherence to TDF/FTC preexposure prophylaxis for HIV prevention in MSM in Los Angeles, California. J Acquir Immune Defic Syndr 2017; 76(5): 501–511.

World

Health Organization

. Consolidated guidelines on HIV prevention, Testing, Treatment, service delivery and monitoring: recommendations for a public health approach [Internet]. WHO.int, 2022 [cited 2022 Jan 16]; Available from: https://www.who.int/publications-detail-redirect/9789240031593

Da Silva

Rojo

Naldinho

, et al. HIV drug resistance prevention and response in the light of the WHO HIV drug resistance report 2021 [Internet]. WHO.int, 2022 [cited 2022 Jan 16]; Available from: https://www.who.int/news-room/events/detail/2021/11/24/default-calendar/hiv-drug-resistance-report-2021-webinar

Koenig

Mounzer

Daughtridge

, et al. Urine assay for tenofovir to monitor adherence in real time to tenofovir disoproxil fumarate/emtricitabine as pre-exposure prophylaxis. HIV Med 2017; 18(6): 412–418.

Lalley-Chareczko

Hiserodt

Moorthy

, et al. Urine assay to measure tenofovir concentrations in patients taking tenofovir alafenamide. Front Pharmacol 2020; 11: 286.

Daughtridge

Hebel

Hashim

, et al. Development and validation of a point-of-care, urine assay to measure adherence to PrEP and ART. International AIDS Society Conference, 2019. [cited 2022 Jan 26]; Available from: http://programme.ias2019.org/Abstract/Abstract/4773

10.

Drain

Kubiak

Siriprakaisil

, et al. Urine tenofovir concentrations correlate with plasma and relate to tenofovir disoproxil fumarate adherence: a randomized, directly observed pharmacokinetic trial (TARGET Study). Clin Infect Dis 2019; 70(10): 2143–2151.

11.

The Joint United Nations Programme on HIV/AIDS . UNAIDS lesotho country data 2019 [Internet]. Unaids.org, 2022 [cited 2022 Jan 16]; Available from: https://www.unaids.org/en/regionscountries/countries/lesotho

12.

Government

of Lesotho

. National guidelines on the use of antiretroviral therapy for HIV prevention and treatment. 9th ed. [Internet]. Children and AIDS, 2022 [cited 2022 Jan 16]; Available from: https://www.childrenandaids.org/. https://www.childrenandaids.org/sites/default/files/2017-04/Lesotho_ART-Guidelines_2016.pdf.

13.

Government of Lesotho . Addendum to the national guidelines on the use of antiretroviral therapy for HIV prevention and treatment. Ministry of Health, Government of Lesotho, 2019.

14.

Brown

Nsakala

Mokhele

, et al. Viral suppression after transition from nonnucleoside reverse transcriptase inhibitor- to dolutegravir-based antiretroviral therapy: a prospective cohort study in Lesotho (DO-REAL study). HIV Med 2022; 23(3): 287–293.

15.

Podsadecki

Vrijens

Tousset

, et al. “White coat compliance” limits the reliability of therapeutic drug monitoring in HIV-1-infected patients. HIV Clin Trials 2008; 9(4): 238–246.

16.

Sevenler

Niu

Dossantos

, et al. Validation of rapid semiquantitative lateral flow assay for Urine tenofovir [CROI abstract 352]. abstracts from the virtual 2021 conference on retroviruses and opportunistic infections. Top Antivir Med 2021; 29(1): 122–123.

17.

Niu

Sevenler

Dossantos

, et al. Implication of measuring Urine tenofovir by rapid lateral flow assay for dose recency [CROI abstract 353]. abstracts from the virtual 2021 conference on retroviruses and opportunistic infections. Top Antivir Med 2021; 29(1): 122–123.

18.

Jennings

Kellermann

Spinelli

, et al. Drug resistance, rather than low tenofovir levels in blood or Urine, is associated with tenofovir, emtricitabine, and efavirenz failure in resource-limited settings. AIDS Res Hum Retroviruses 2021. doi: 10.1089/AID.2021.0135.

19.

Pepperrell

Venter

WDF

McCann

, et al. Participants on dolutegravir resuppress human immunodeficiency virus RNA after virologic failure: updated data from the advance trial. Clin Infect Dis 2021; 73(4): e1008–e1010.