Short Communication: A Descriptive Analysis of Dried Blood Spot Adherence Testing Among Ugandans with HIV Presenting with Cryptococcal Meningitis

Abstract

Early antiretroviral therapy (ART) initiation after cryptococcal meningitis increases mortality, and those unmasking cryptococcosis after <2 weeks of ART have higher mortality. However, it is unknown if those presenting as ART experienced are actually adherent to their ART. Unknowingly, restarting ART in persons, who have discontinued ART, may be a fatal iatrogenic error. To evaluate ART adherence in an exploratory analysis, we collected dried blood spots on 44 HIV-infected persons presenting with cryptococcal meningitis. We quantified tenofovir diphosphate (TFV-DP) and lamivudine (3TC) from dried blood spots. We quantified cumulative ART adherence over the preceding 6–8 weeks based on TFV-DP concentrations and adherence over the last few days based on 3TC concentrations. Of 22 ART experienced, 20 (91%) had quantifiable concentrations. Of 18 receiving tenofovir, 15 (83%) had TFV-DP consistent with drug intake of ≥4 doses/week or moderate adherence. With 3TC, 72% (18/22) had detectable levels consistent with adherence over the last 3 days before measurement. Only three ART-experienced subjects were alive and virally suppressed at 4 months (n = 2 on ART for <30 days; n = 1 with undetectable antiretrovirals). Surprisingly, of 22 who reported not receiving ART, 4 (18%) had quantifiable tenofovir. Most ART-experienced subjects were taking their ART with moderate to good adherence with the majority likely having viral resistance given generally at good ART levels, receipt of intensive adherence counseling, and lack of subsequent viral suppression. The World Health Organization (WHO) guidelines recommend adherence counseling with ART continuation and repeat viral loads in 1–3 months before switching to second-line ART. These recommendations are likely inappropriate in those with central nervous system infections given the additional possible harm of central nervous system immune reconstitution syndrome. Further study to evaluate continuation of ART regimens when presenting with cryptococcosis has benefit, with checking blood levels at presentation potentially being a helpful option. ClinicalTrials.gov Identifier: NCT01802385.

Background

Earlier antiretroviral therapy (ART) initiation after cryptococcal meningitis causes increased mortality compared to deferring ART initiation for 4–6 weeks.¹ Those initiating ART within 30 days (especially within 14 days) antecedent to presentation with cryptococcosis also have twofold higher mortality than those on ART >30 days or ART-naive.² However among those presenting ART experienced with suspected virologic failure, AIDS progression, and cryptococcal meningitis, it is often unknown if patients are actually adherent to ART. While the default practice is to continue ART, restarting ART at hospital admission in a patient who has discontinued their ART could be detrimental.^1,2

Prior studies have shown varied correlation between self-reported ART adherence and actual adherence with self-report often being overstated.^3

–7 Many situational, patient-specific, and methodological factors are associated with different degrees of correlation between self-report and actual adherence.^6,8 To date, studies on ART adherence in Africa have been conducted in outpatient settings. In contrast, patients presenting to hospital with meningitis are often quite ill and frequently confused, further complicating interpretation of self-reported adherence.

Management of ART in people presenting with meningitis is therefore quite challenging. Current World Health Organization (WHO) guidelines recommend continuing ART in those with first episode of detectable viral load and recommend intensive adherence counseling with then rechecking a viral load in 1 to 3 months.⁹ However, it is complex in people with cryptococcal meningitis. If people with HIV have been nonadherent, but reinitiate ART shortly after presenting with meningitis, immune reconstitution inflammatory syndrome (IRIS) can cause increased mortality.¹ However, stopping ART could induce resistance and potentially send an unintended psychological message to the individuals about the need for their HIV medications. To better guide treatment, we evaluated ART adherence in those presenting with cryptococcal meningitis by quantitative measurement of tenofovir diphosphate (TFV-DP) and lamivudine-triphosphate (3TC-TP) using dried blood spots.

Methods

In this exploratory descriptive analysis we collected dried blood spots on sequential HIV-infected persons presenting with cryptococcal meningitis from August to December 2016 in Kampala, Uganda to evaluate ART adherence. This evaluation of ART adherence was a substudy from a larger meningitis trial.¹⁰ All participants or their surrogate provided written informed consent. IRB approval occurred at Mulago Research Ethics Committee in Uganda and the University of Minnesota.

We measured TFV-DP, which quantifies cumulative ART adherence and exposure over the preceding 6–8 weeks due to its 17-day half-life in this matrix. Second, we measured 3TC-TP, which we estimate has a half-life of ∼35 h in blood spots, similar to emtricitabine-triphosphate (FTC-TP).¹¹ Quantifiable 3TC-TP would represent a dose within the preceding 2–3 days. Dried blood spots were collected on the day of hospital presentation and processed at the University of Colorado as previously described.^12
–14 Briefly, a 3 mm punch was extracted from the blood spot card in 70:30 methanol-water. TFV-DP and 3TC-TP were quantified using liquid chromatography–tandem mass spectrometry as previously described.^11,12,14 Lower limits of quantitation were 25 fmol/sample for TFV-DP and 0.1 pmol/sample for 3TC-TP.

Demographic, adherence (self-reported or caregiver-reported), and clinical data were prospectively collected. Statistical analysis was primarily descriptive. Sample size was driven by budgetary constraints.

We compared drug concentrations by ART status as either ART experienced or ART naive as per self-report. As many participants presented with altered mental status at baseline with initially unknown ART status, we collected samples from all participants. Those on ART had their ART continued if clinically stable and taking oral medications. All participants (or caregivers if the participant was confused) received intensive adherence counseling during their hospitalization by a trained counselor. We measured HIV RNA in surviving participants on ART as per clinician discretion. Altered mental status was defined using a Glasgow Coma Scale (GCS) <15. In these critically ill individuals, often with vomiting, headaches, dizziness, and/or fevers, the GCS offered a quick way to evaluate mentation when a more thorough evaluation such as the Montreal Cognitive Assessment (MoCA) test was impractical for patients with acute delirium or coma.

Adherence interpretation

We utilized previously published cutoffs to interpret adherence to tenofovir disoproxil fumarate (TDF)-based ART (1,250 fmol/punch = 7 doses/week; 700–1,249 fmol/punch = 4–6 doses/week; 350–699 fmol/punch = 2–3 doses/week; and <350 fmol/punch = <2 doses/week),¹⁵ which have also been associated with the odds of HIV viral suppression.¹⁶ In subjects who had started ART <6 weeks, cutoffs were adjusted based on accumulation shown by Anderson et al. using published medians at 2 and 4 weeks, multiplying by 1/(1 − e^−0.04)*days on the drug.¹² The 3TC-TP concentrations were interpreted in a binary manner as quantifiable versus below the limit of quantification to establish 3TC receipt over the past ∼3 days.^14,17

Results

We prospectively collected dried blood spots from 44 Ugandans with cryptococcal meningitis. This was a subset of the entire 460 subject cohort of whom 223 reported being on ART.¹⁰ Of the 216 reporting which ART they were on, 216 (100%) subjects were on lamivudine (3TC)-based ART in fixed-dose combination tablets with another nucleoside reverse transcriptase inhibitor (NRTI). The second NRTI was: TDF in 177 (81.5%), zidovudine (AZT) in 38 (17.6%), and abacavir (ABC) in 1 (0.5%) participant (one individual was on 3TC/TDF/ABC). The third antiretroviral was most frequently efavirenz in 155 (76.7%), nevirapine in 30 (13.9%), boosted atazanavir in 17 (7.9%), lopinavir in 4 (1.9%), ABC in 1 (0.5%) (receiving 3TC/TDF/ABC), and no third antiretroviral medication in 9 (4.2%) participants.

Of the 44 participants with dried blood spots, 20 (45%) were women (Table 1). Median age was 33 years old [interquartile range (IQR), 28–37]. Median CD4 count was 12 (IQR, 5–56) cells/μL. Median weight was 52 kg (IQR, 50–60). Twenty-two reported being ART experienced with a median duration of 8 months (IQR, 2–31) with three subjects initiating ART within 30 days. None of the participants had IRIS at presentation or in the 3 months subsequent. Four participants (9%) had sterile cryptococcal cultures with positive cerebrospinal fluid cryptococcal antigen. Four of the participants in this substudy died within 2 weeks of enrollment, one reportedly on ART >6 months and the others ART naive but three of four with some ART in their blood, including the one ART experienced. Ten more participants died in the following 16 weeks, one reportedly on ART for 2 weeks–6 months duration and two with >6 months duration of ART. None of these 10 had TFV in their blood, but those on ART >6 months did have 3TC. This totals to 14 deaths within 18 weeks, consistent with the overall cohort.¹⁰

Table 1.

Baseline Demographics of Individuals with Cryptococcal Meningitis and Baseline Dried Blood Spot Testing Completed

	n (%) or median (IQR)
Age (years)	33 (28–37)
Female	20 (45%)
Weight (kg)	52 (50–60)
Abnormal Baseline Glasgow Coma Score, percent <15	24 (55%)
Baseline CD4 (cells/uL)	12 (5–56)
On HIV therapy	22 (50%)
Duration of HIV therapy
ART naive	22 (50%)
<2 weeks	2 (5%)
2 weeks–6 months	7 (16%)
>6 months	13 (30%)
ART type at baseline of those on therapy
EFV/3TC/TDF	13 (59%)
AZT/3TC/TDF	2 (9%)
NVP/3TC/TDF	1 (5%)
AZT/3TC	3 (14%)
3TC/AZT/NVP	1 (5%)
TDF/3TC/ATV/r	1 (5%)
Baseline opening pressure (cm H20)	31 (20–48)

3TC, lamivudine; ART, antiretroviral therapy; ATV, Atazanavir boosted with Ritonavir; AZT, zidovudine; EFV, Efavirenz; IQR, interquartile range; NVP, Nevirapine; TDF, tenofovir disoproxil fumarate.

ART quantification in participants with current ART use

Of 22 ART-experienced participants, 18 reported receiving tenofovir and 4 AZT but not TDF with three on both (TDF/3TC/AZT). Among the 18 subjects reported as receiving TDF and 3TC, the estimated adherence and virologic responses are presented in Table 2. Three subjects had initiated ART <30 days before presentation of meningitis, while 15 subjects had been taking ART >30 days. Overall, 50% (9/18) had excellent adherence consistent with probable 7 doses/week over the prior 8 weeks, and 17% (3/18) had minimal ART adherence consistent with <2 doses/week of TDF. Overall, of 18 reporting taking TDF, 14 (78%) had TFV-DP concentrations consistent with 4–7 doses/week (median drug concentration 1,316 fmol/punch IQR 942-1872).

Table 2.

Findings from 22 ART Experienced Participants with Dried Blood Spots Tested for TDF and 3TC

	Estimated TDF doses/week	N	3TC-TP quantifiable	Week 1 Viral Load copies/mL	Month 2–6 Viral Load copies/mL	GCS <15	Alive at 4 months
TDF <30 days	4–7	2	2	N/A	<100 (N = 1) N/A (N = 1)	1	2
TDF <30 days	2–3	1	1	N/A	N/A	1	1
TDF 30–180 days	7	1	1	N/A	N/A	1	1
	4–6	3	2	N/A	N/A	2	3
	<2	1	1	<100	<100	1	1
TDF >180 days	7	5	5	>10,000 (N = 2) N/A (N = 3)	>10,000 (N = 3) N/A (N = 2)	0	5
	4–6	3	1	>10,000 (N = 2) N/A (N = 1)	>10,000 (N = 1) Dead (N = 2)	3	1
	<2	2	0	>10,000 (N = 1) <100 (N = 1)	>10,000 (N = 1) <100 (N = 1)	0	2
AZT >180 days	N/A not on TDF	3	0	N/A	Dead (N = 3)	2	0
AZT 30–180 days	N/A not on TDF	1	0	N/A	>10,000	1	0

Viral loads were checked as per clinician request and allowed by national HIV testing guidelines.

3TC-TP, lamivudine triphosphate; GCS, Glasgow Coma Scale.

In the majority of subjects, the reported adherence and corresponding drug concentrations were aligned. Of the four participants receiving AZT, none had quantifiable TFV-DP, as expected, and three had quantifiable 3TC-TP, suggesting recent ART use. Of 22 reporting to be taking 3TC, 16 (73%) had quantifiable 3TC-TP, implying recent ART use in the 2–3 days before hospital admission. We only had viral load testing on eight individuals 2 or more weeks after enrollment. None of the five on ART >6 months was virally suppressed. Three of those five had TFV-DP concentrations consistent with probably 7 doses/week in the prior 8 weeks. These three had detectable 3TC. Of the other two individuals, one had TFV-DP concentrations consistent with 4–7 doses/week and the other had consistent with <2 doses/week of TDF with neither having detectable 3TC concentrations.

ART quantification in ART-naive participants

Overall, 22 participants reported being ART-naive. However, of those 22 ART-naive participants, 4 (18%) had quantifiable antiretroviral drug concentrations. Two of these individuals died within 2 weeks. The first had TFV-DP concentration reflecting <2 doses/week and unquantifiable 3TC-TP. The second had quantifiable 3TC-TP but not TFV-DP. The remaining two participants had both quantifiable TFV-DP and 3TC-TP, consistent with recent dosing and cumulative adherence of >4 doses/week. Three out of four of the individuals with discordant self-reported adherence and drug concentrations had altered mental status with a GCS <15 at baseline suggesting likely unreliable history. One of the four with GCS of 15 had TFV-DP concentrations consistent with excellent adherence and quantifiable 3TC-TP.

Associations with adherence

Regarding the association of altered mental status with adherence, 53% (8/15) of those with TFV-DP concentrations consistent with moderate to excellent adherence had a GCS <15. This percentage is in contrast to 33% (1/3) individuals with TFV-DP concentrations consistent with poor adherence (<3 doses/week). Of the 12 with quantifiable 3TC-TP, 5 (42%) had a GCS <15 compared with the 6 (50%) with undetectable 3TC-TP, none of whom had an abnormal GCS.

Discussion

Our descriptive analysis of a limited number of samples shows that most participants who reported receiving ART had drug concentrations consistent with their report. A higher percentage of subjects had moderate to excellent adherence by TFV-DP (83%) than quantifiable 3TC-TP (73%), which suggests better adherence over the prior 6–8 weeks than the prior few days. This discrepancy could be consistent with a patient unable to take oral medications as they developed nausea, emesis, and confusion from meningitis. Conversely, as patients became ill with a subacute progression of cryptococcosis over multiple weeks, patients may have restarted ART in an attempt to improve their health.

These data suggest that self-reported adherence is consistent with drug concentrations. We found this surprising because while ART adherence in Africa is generally high,¹⁸ individuals presenting with cryptococcal meningitis generally have low CD4 counts often consistent with poor ART adherence. Our data suggest that individuals presenting with cryptococcal meningitis had moderate-excellent to ART. However, a larger clinical problem exists in knowing how best to manage those presenting with cryptococcosis on ART. We previously reported that people who have started ART within 14 days before meningitis presentation have nearly twofold higher risk of death than those ART-naive or receiving ART for more extended periods.² Similarly, early initiation of ART after cryptococcal diagnosis within 2 weeks causes harm.¹ Whether stopping ART in those recently initiated on ART mitigates the excess risk of death remains unknown. Data to guide medical decision-making are lacking, and clinical trials are urgently needed. Risks of continuing ART after presenting with suspected meningitis include (1) potential drug–drug interactions coupled with the likelihood that individuals may have virologic resistance to their ART, and (2) the possibility that the individuals were ART nonadherent, and thus, the medical team is actually reinitiating ART and increasing risk for IRIS and death. Our data herein suggest that <30% of individuals were nonadherent to ART at the time of cryptococcosis diagnosis.

Given that second-line ART is more expensive and less well tolerated than first line, and switching eliminates future treatment options from patients with a lifelong disease,¹⁹ measuring drug levels to check ART adherence may be cost-effective, especially in those with confusion or lacking documentation. ART drug concentrations could help differentiate between those not taking their ART who could use adherence counseling versus those who should be started on second line as soon as possible. Since drug testing can now be done on dried blood spots with minimal processing requirements (although −80°C storage is required),¹¹ this could be a feasible option in some areas of sub-Saharan Africa.²⁰

There are limitations to this analysis, mainly due to the descriptive nature. Analysis of ART adherence through dried blood spot testing was substudy of a larger clinical trial, and therefore, the substudy was not powered to perform robust statistical analysis. However, we believe that these data provide critical novel insight into adherence patterns of those presenting with opportunistic infections, which is the next crucial step in the development of a thoughtful approach on how to handle ART in this population. Further research would be needed to confirm these descriptive results.

Conclusion

Over three-quarters of ART-experienced subjects presenting with cryptococcal meningitis were taking their ART with moderate to excellent adherence. The majority of those ART experienced and on ART >30 days likely had viral resistance given drug concentrations consistent with good adherence, intensive adherence counseling, and lack of viral suppression. However, a few had drug concentrations consistent with nonadherence but became virally suppressed when they were restarted on ART. TFV-DP and 3TC-TP concentrations in dried blood spots may help differentiate between viral resistance and nonadherence in these individuals with cryptococcal meningitis and help guide management.

Footnotes

Acknowledgments

The authors thank all members of the Adjunctive Sertraline for the Treatment Of Cryptococcal Meningitis (ASTRO-CM) study team for their contributions to this work. The authors also thank all of the participants in the study.

Authors' Contributions

S.M.L. is the guarantor of integrity of entire study. S.M.L. and D.R.B. defined intellectual content. S.M.L., T.K.K., J.C.-M., P.L.A., E.M., L.T., K.S., J.R., and D.B.M. were involved in data acquisition. S.M.L., M.R.N., J.C.-M., A.S.B., K.H.H., and D.R.B. were primarily involved in data analysis. S.M.L., M.R.N., and D.R.B. were primarily involved in article preparation. All authors read and approved the final article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported, in part, by the National Institute of Neurologic Diseases and Stroke (NINDS, R01NS086312), Fogarty International Center (R25TW009345), the National Institute of Allergy and Infectious Diseases (T32AI055433, K08AI134262-02), and the National Institute of Mental Health (K23MH121220).

References

Boulware

, Meya

, Muzoora

, et al.: Timing of antiretroviral therapy after diagnosis of cryptococcal meningitis. N Engl J Med, 2014; 370:2487–2498.

Rhein

, Hullsiek

, Evans

, et al.: Detrimental outcomes of unmasking cryptococcal meningitis with recent ART initiation. Open Forum Infect Dis, 2018; 5:ofy122.

Kabore

, Muntner

, Chamot

, Zinski

, Burkholder

, Mugavero

: Self-report measures in the assessment of antiretroviral medication adherence: Comparison with medication possession ratio and HIV viral load. J Int Assoc Provid AIDS Care, 2015; 14:156–162.

Wilson

, Carter

, Berg

: Improving the self-report of HIV antiretroviral medication adherence: Is the glass half full or half empty?. Curr HIV/AIDS Rep, 2009; 6:177–186.

Haberer

, Bwana

, Orrell

, et al.: ART adherence and viral suppression are high among most non-pregnant individuals with early-stage, asymptomatic HIV infection: An observational study from Uganda and South Africa. J Int AIDS Soc, 2019; 22:e25232.

Nieuwkerk

, Oort

. Self-reported adherence to antiretroviral therapy for HIV-1 infection and virologic treatment response: A meta-analysis. J Acquir Immune Defic Syndr, 2005; 38:445–448.

Thirumurthy

, Siripong

, Vreeman

, et al.: Differences between self-reported and electronically monitored adherence among patients receiving antiretroviral therapy in a resource-limited setting. AIDS, 2012; 26:2399–2403.

Ingle

, Crane

, Glass

, et al.: Identifying risk of viral failure in treated HIV-infected patients using different measures of adherence: The antiretroviral therapy cohort collaboration. J Clin Med, 2018; 7:328.

World Health Organization: Consolidated Guidelines on The Use of Antiretroviral Drugs for Treating and Preventing HIV Infection. Recommendations for the Public Health Approach, 2nd Ed. WHO, Geneva, 2016. Available at http://www.who.int/hiv/pub/arv/arv-2016/en/, accessed July 1, 2020 .

10.

Rhein

, Huppler Hullsiek

, Tugume

, et al.: Adjunctive sertraline for HIV-associated cryptococcal meningitis: A randomised, placebo-controlled, double-blind phase 3 trial. Lancet Infect Dis, 2019; 19:843–851.

11.

Castillo-Mancilla

, Zheng

, Rower

, et al.: Tenofovir, emtricitabine, and tenofovir diphosphate in dried blood spots for determining recent and cumulative drug exposure. AIDS Res Hum Retroviruses, 2013; 29:384–390.

12.

Anderson

, Liu

, Castillo-Mancilla

, et al.: Intracellular tenofovir-diphosphate and emtricitabine-triphosphate in dried blood spots following directly observed therapy. Antimicrob Agents Chemother, 2018; 62:e01710-17.

13.

Bushman

, Kiser

, Rower

, et al.: Determination of nucleoside analog mono-, di-, and tri-phosphates in cellular matrix by solid phase extraction and ultra-sensitive LC-MS/MS detection. J Pharm Biomed Anal, 2011; 56:390–401.

14.

Zheng

, Rower

, McAllister

, et al.: Application of an intracellular assay for determination of tenofovir-diphosphate and emtricitabine-triphosphate from erythrocytes using dried blood spots. J Pharm Biomed Anal, 2016; 122:16–20.

15.

Brooks

, Anderson

: Pharmacologic-based methods of adherence assessment in HIV prevention. Clin Pharmacol Ther, 2018; 104:1056–1059.

16.

Castillo-Mancilla

, Morrow

, Coyle

, et al.: Tenofovir diphosphate in dried blood spots is strongly associated with viral suppression in individuals with human immunodeficiency virus infections. Clin Infect Dis, 2019; 68:1335–1342.

17.

Frasca

, Morrow

, Coyle

, et al.: Emtricitabine triphosphate in dried blood spots is a predictor of viral suppression in HIV infection and reflects short-term adherence to antiretroviral therapy. J Antimicrob Chemother, 2019; 74:1395–1401.

18.

Bijker

, Jiamsakul

, Kityo

, et al.: Adherence to antiretroviral therapy for HIV in sub-Saharan Africa and Asia: A comparative analysis of two regional cohorts. J Int AIDS Soc, 2017; 20:21218.

19.

World Health Organization: Updated recommendations on first-line and second-line antiretroviral regimens and post-exposure prophylaxis and recommendations on early infant diagnosis of HIV: Interim guidelines. Supplement to the 2016 consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection. WHO, Geneva, 2018.

20.

Lofgren

, Morrissey

, Chevallier

, et al. Evaluation of a dried blood spot HIV-1 RNA program for early infant diagnosis and viral load monitoring at rural and remote health care facilities. AIDS, 2009; 23:2459–2466.