HIV regimen adherence Lessons for Long Acting IM injectables from - - PowerPoint PPT Presentation

HIV regimen adherence

Lessons for Long Acting IM injectables from oral ART & PrEP

Measures to improve ART & PrEP adherence

• Inherent drug improvement to increase tolerability
• Single regimen ART
• Fewer side effects
• Improved delivery method
• Long acting intramuscular injectables
• Oral ART pharmacy refills for longer periods
• Monitored adherence support
• Objective technological measures (EAMs, RNA & drug levels)
• Subjective personal measures (self-report, caregiver report, SMS)
• Person-centred approaches using both/either of above

Concern – if oral ART adherence is difficult monitor, what lessons can be learnt to improve chance of success for adherence to infrequent LA IMs injectables?

Review of published studies

• Search of published studies using PubMed
• Papers between 2009 – 2017
• Searched for ‘adherence’ and;
• SMS, short message service
• MEMS, MEMS-cap, EAM
• Peer support, peer to peer, counselling, support
• Economic incentive, cash incentive, voucher
• HIV RNA test, plasma drug levels, hair
• Filtered for HIV studies (PrEP & ART)
• Reviewed 57 studies

SMS & Instant messaging EAMs (MEMS-cap) Real-time wireless EAMs Drug levels (plasma or hair) HIV RNA testing Digital medicine systems Barriers to success, e.g. Self-report bias, no airtime Pocket dosing, curiosity openings Mobile network loss, battery life Dose masking (delayed testing) Cost & equipment requirements Failure of

• ne/more tech

components

Combinations of measures to cross-validate adherence & develop intelligent testing procedures based on individual’s behaviour patterns

Technological measures of adherence

SMS – Surveys to understand adherence behaviour

Haberer et al 20171:

• N = 373, Uganda
• Daily SMS surveys - monitoring

sexual behaviour & PrEP use in serodiscordant couples  Mean reported PrEP adherence: 92% on surveys concurrently reporting sex, 84% on surveys reporting no sex  Higher PrEP adherence associated with increased HIV risk

FIGURE 1. Mean PrEP adherence as associated with risk for HIV transmission. Circles indicate risk behaviors for HIV acquisition: <6 months of ART use by the partner living with HIV, sex reported within the serodiscordant partnership, and reported condomless sex. Mean reported PrEP adherence concurrent with each overlap of behaviors is shown in the legend. Haberer et al 2017.

SMS – Reminders during adherence lapses

Sabin et al 20152:

• N = 120, China
• Individualised SMS reminders sent

to patients if dose-taking was late by 30+ mins (recorded from MEMS data)

• MEMS-generated data also used

in counselling for ‘late’ patients  Improve ART adherence (93.3%) vs control group (no SMS or counselling) (84.7%)

FIGURE 2. Monthly mean adherence among intervention and control subjects, stratified by pre-intervention period optimal (≥95%) or suboptimal (<95%) adherence, using an on-time adherence measure Note: Pre-intervention period refers to Months 1–3; intervention period is the subsequent 6-month period (Months 4–9) during which subjects received triggered reminders and data-informed counselling. Sabin et al 2015.

Electronic Adherence Monitoring – Understanding the difference between self-report & MEMS adherence

Musinguzi et al 20163:

• N = 1147, 6048 person-months,

Uganda

• Self-reported PrEP adherence was

compared with MEMS data, unannounced pill counts & TDF plasma levels (N = 365, SSDD: 74%)  SSDD vs <SSDD levels was poor for self-reported adherence (AROC 0.54 – 0.56) & UPC (AROC 0.58), but moderate for MEMS monitoring (AROC 0.70).

FIGURE 3. Receiver–operating curves for the sensitivity of self-reported adherence, unannounced pill counts, and electronic monitoring to detection of steady-state daily dosing plasma tenofovir levels. EM, electronic monitoring; SR, self-reported; SSDD, steady-state daily dosing; UPC, unannounced pill counts. Musinguzi et al 2016.

Electronic Adherence Monitoring – Real-time wireless monitoring to understand risk of viral rebound

Haberer et al 20154:

• N = 479, Uganda
• Measured interruptions to

adherence using real-time wireless EAMs; interruptions 48+ hours followed up with HIV RNA testing  Odds of viral rebound increased by 25% with each day beyond 48 hours (OR: 1.25; P = 0.007)  Viral rebound was also associated with 30-day adherence before the interruption (OR: 0.73; P = 0.02).

FIGURE 4. Association between duration of adherence interruption and viral rebound. A total of 587 interruptions were seen among 261 of the total 479 participants (54%). Viremia exceeded 5% of interruptions lasting longer than 7

• days. Twenty-one interruptions (3%) were longer than 14 days, none of which had detectable HIV RNA during the

interruption or at the subsequent routine quarterly assessment. Haberer et al 2015.

HIV RNA testing – MEMS-based predictive testing

Petersen et al 20155:

• N = 1478, 16 sites in USA
• Applied machine learning algorithm

to MEMS data, CD4+ T cell counts & ART regimen  Improved classification of virological failure from MEMS, to predict selective RNA testing  25–31% of HIV RNA tests could be avoided  Cost savings: $16-$29/person-mth

FIGURE 5: Cross-validated ROC Curves for classification of virological failure using four Super Learner prediction models and three-month “average adherence” (percent prescribed doses recorded by MEMS). Failure defined as HIV RNA level>400 copies/ml. Sensitivity for failure detection ≥95%. Petersen et al 2015.

HIV RNA testing – new technologies for POC tests

Gurrala et al 20166:

• Development of a novel chip based,

point-of-care HIV-1 viral load assay; amplifies & detects HIV-1 RNA

• N = 991 clinical samples

 Sensitivity: 95% (in vitro) & 88.8% (on- chip) at >1000 RNA copies/reaction  Median time to detection: 20.8 mins  Sensitivity, specificity & reproducibility close to that required for a low-power point-of-care device e.g. USB stick

FIGURE 6: (a) Image of prototype chip for amplification and detection of nucleic acids compatible with a USB port. (b) Schematic of a

• chip. Each chamber functions independently,

when the pH of the chamber changes the ISFET (ion sensitive field effect transistor) generates an electrical signal. Gurrala et al 2016.

Technological measures of adherence Economic incentives Peer support Culturally sensitive support Person-centred support through understanding what matters to individuals in order to motivate adherence using most appropriate technological measures

Combining technological measures & meaningful support

Economic Incentives

El Sadr 20177 & Greene 20178:

• N = 1061, 18 test sites, USA
• Offered vouchers for linkage to care ($125), &

quarterly vouchers conditional on viral suppression ($70)  Incentives did not increase linkage to care (AOR: 1.10; 95%CI, 0.73-1.67; P = .65), but significantly increased viral suppression (4.9% higher than control) (95%CI, 1.4%-8.5%; P = .007)  Qualitative follow up (Greene 2017) showed broader positive impact, including patients proactively attending the doctor

El Sadr et al. 2017

Blashill et al 20159:

• N = 331, USA
• Undertook diagnostics of

psychosocial syndemics & monitored adherence to ART using MEMS  Co-occurring psychosocial problems have an additive effect on the risk for poor ART adherence  Integrative cognitive-behavioural interventions to support adherence should be given attention

Understanding personal barriers to adherence

FIGURE 8: Percent Adherence by Syndemic Group. Blashil et all 2015. Psychosocial syndemics are psychosocial issues which occur together in one patient’s life, and can include childhood abuse, current violence, alcohol or substance abuse/dependence, post-traumatic stress disorder, anti-social personality disorder, anxiety spectrum disorders, mood disorders, and psychotic disorders 4.1 times greater

• dds of

being non- adherent

(95% CI: 0.84, 20.4, P = 0.08)

5.0 times greater

• dds of

being non- adherent

(95% CI: 1.02, 24.4, P = 0.047)

8.5 times greater

• dds of

being non- adherent

(95% CI: 1.7, 42.9, P = 0.01)

Odds compared to zero syndemics

Person-centred, nurse-led monitoring

De Bruin et al 201710:

• N = 221, Netherlands
• AIMS is a nurse-led 1-2-1 behavioural

intervention, incorporating pre- & during- study MEMS data, for adherence goal setting  At 5, 10 & 15 months log viral load was 1·26 times higher (95% CI 1·04–1·52) in the control group than in AIMS group  AIMS was cost-effective: reduced lifetime societal costs by €592/patient & increased QALYs by 0·034/patient

Culturally sensitive peer support

Bogart et al 201711:

• N = 215, USA
• Peer-delivered MEMS-informed

counselling, adapted for HIV+ African Americans; “Rise”

• Addressed culturally congruent

adherence barriers (e.g., medical mistrust, HIV stigma) & assisted with linkage to supportive services  Adherence improved relative to control group, with large cumulative effect after 6 mths (OR: 4.76)

• Fig. 2 Adherence and nonadherence patterns from baseline to 6-month follow-up. Bogart et al 2017.

Adherence improved over time relative to control group, (OR = 1.30 per month (95% CI = 1.12–1.51, p < 0.001), representing large cumulative effect after 6 mths (OR = 4.76, Cohen’s d = 0.86).

Co-designed, peer-led support interventions

Enriquez 201512:

• N = 20, USA
• Adapted a nurse-led Wellness Motivation

approach to a peer-led counselling format through co-design with lay peers

• Peer-led group attended 100% of sessions,

control group attended 60%  Peer-led group had significantly improved adherence at 12 and 24 weeks  Suppression of viral load correlated with higher MEMS events & higher number of on- time pharmacy refills.

12 week post viral load log drop 24 week post viral load log drop ‘Ready’ peer-led adherence intervention M (SD) (n =10) 2.6180 (0.8113) 3.3439 (1.1101) ‘Healthy Eating’ control intervention M (SD) (n=10) 0.7036 (1.1492) 0.6143 (1.8760) Difference M (SD) 1.9144 (0.9947) 2.7295 (1.5414) Test statistics

• 4.30
• 3.96

P 0.0004 0.0009

HIV RNA PCR log drop: baseline to 12 & 24 weeks post-intervention. Enriquez et

• al. 2015.

Lessons for Long Acting IM injectables?

Cross validating adherence measures  overcome barriers to monitoring adherence $ Costs of tech reducing Feasibility increasing + Qualitative data from patients (what’s meaningful to them) Behavioural adherence data + Long term funding commitments Time to market for Long Actings can be used to develop these approaches & secure funding commitments 2020 - 2025 Intelligent, timely use of behavioural data to: send reminders, & reduce RNA testing requirements & cost + Meaningful, culturally- sensitive adherence support delivered by nurses &/or peers

References

1. Haberer, J. et al. Real-time electronic adherence monitoring plus follow-up improves adherence compared with standard electronic adherence

• monitoring. AIDS 2017, 31:169–173.

2. Sabin, LL. et al. Improving adherence to antiretroviral therapy with triggered real-time text message reminders: the China adherence through technology study. J Acquir Immune Defic Syndr 2015; 69:551–559. 3. Musinguzi, N. et al. Comparison of subjective and objective adherence measures for preexposure prophylaxis against HIV infection among serodiscordant couples in East Africa. AIDS 2016, 30:1121–1129. 4. Haberer, J. et al. Duration of Antiretroviral Therapy Adherence Interruption Is Associated With Risk of Virologic Rebound as Determined by Real- Time Adherence Monitoring in Rural Uganda. J Acquir Immune Defic Syndr 2015; 70:386-392. 5. Petersen, M. et al. Super learner analysis of electronic adherence data improves viral prediction and may provide strategies for selective HIV RNA

• monitoring. J Acquir Immune Defic Syndr. 2015 May 1; 69(1): 109–118.

6. Gurrala, R. et al. Novel pH sensing semiconductor for point-of-care detection of HIV-1 viremia. Scientific Reports 2016; 6:36000. 7. El Sadr, W. et al. Financial Incentives for Linkage to Care and Viral Suppression Among HIV-Positive Patients. A Randomized Clinical Trial (HPTN 065). JAMA Intern Med. 2017;177(8):1083-1092. 8. Greene, E. et al. “It Makes You Feel Like Someone Cares”; acceptability of a financial incentive intervention for HIV viral suppression in the HPTN 065 (TLC-Plus) studyPLoS ONE 12(2): e0170686. 9. Blashill, A. et al. Psychosocial Syndemics are Additively Associated with Worse ART Adherence in HIV-infected Individuals. AIDS Behav. 2015 June ; 19(6): 981–986. 10. De Bruin, M. et al. Effectiveness and cost-effectiveness of a nurse-delivered intervention to improve adherence to treatment for HIV: a pragmatic, multicentre, open-label, randomised clinical trial. Lancet Infect Dis 2017; 17: 595–604. 11. Bogart, L. et al. A Randomized Controlled Trial of Rise, a Community-Based Culturally Congruent Adherence Intervention for Black Americans Living with HIV. Ann. behav. med. (2017) 51:868–878. 12. Enriquez, M. et al. A Peer-Led HIV Mediation Adherence Intervention Targeting Adults Linked to Medical Care but without a Suppressed Viral

• Load. J Int Assoc Provid AIDS Care. 2015 ; 14(5): 441–448.