Highly active antiretroviral therapy (HAART) transformed a once fatal condition into a chronic, manageable condition. However, it is estimated that 20-40% of patients on 2nd line treatment (2 nucleotide reverse transcriptase inhibitors [NRTIs] and a boosted protease inhibitor [PI]) are failing treatment. Figures are thought to be higher in children and adolescents.

The reason why patients are failing 2nd line treatment is not exactly known. Failure has been previously attributed to poor adherence. However, some literature shows that some patients on boosted PIs achieve and maintain viral suppression despite suboptimal adherence (adherence of 80- 95%). Viral factors, like drug resistance, are also implicated in treatment failure. However, boosted PIs have high genetic barrier to clinically significant mutations. Therefore, a virus would have to harbour multiple PI mutations for the virus to have reduced susceptibility to boosted PI regimens. Pharmacological factors such as suboptimal dosing, impaired absorption and drug interactions may also be responsible for treatment failure.

If sub-optimal adherence is the reason why children are failing 2nd line treatment, then restoring optimal adherence should result in viral suppression, failure of which might mean that other causes are contributing to failure. If resistance is the cause of treatment failure, then this study will provide evidence for advocating for resistance testing and the use of 3rd line antiretroviral drugs. If children with adequate adherence demonstrate inadequate drug levels in their plasma, then this study will provide evidence to advocate for studies to examine reasons for inadequate drug exposure amongst HIV-infected children. These studies are paramount to optimizing dosing algorithms in this population.

This proposed study will help elucidate reasons for treatment failure in HIV-infected children on second line treatment with the aim of ultimately optimizing antiretroviral treatment strategies for this important group.

This is a randomized, controlled, interventional study, which will be conducted in 2 phases.

Phase 1

The first phase of the study will include:

1. Screening for study eligibility (viral loads and administering eligibility screening form
2. Randomisation to study arms (1:1)- random computer generated numbers, sealed in opaque envelopes, clinic nurse keeps the envelopes and produces them sequentially when asked for
3. Enrolment- informed consent and assent, 2 questionnaires administered, hair samples collected
4. Three months of follow-up
5. Post- intervention/control questionnaire administered, viral load and hair samples collected again
6. Patients with viral load <1 000 copies/ml from both arms exit study here. Those with viral load > = 1 000 copies/ml from both arms proceed to phase 2

Phase 2

Patients with viral load > = 1 000 copies/ml from both arms have genotyping for drug resistance

Hair pharmacokinetic (PK) study

Some participants may re-suppress after adherence counselling. Hair anti-retroviral (ARV) levels will show whether the patients have been taking their ARVs or not in the past 3 months. Suboptimal ARV levels in hair may mean that adherence was suboptimal or the dose the patient is taking is suboptimal. Presence of adequate drug levels while the patient is still virologically failing treatment may imply that there is resistance.

Hair collection is being done according to Women's Interagency HIV Study hair collection protocol as described below. All hair samples are stored in a cool, dry place in the department of Clinical Pharmacology in the University of Zimbabwe. Analysis will be done by a validated method at University of Carlifornia San Francisco (UCSF).

Hair collection procedure:

1. Clean the blades of a pair of scissors with an alcohol pad and allow blades to completely dry prior to use
2. Lift up the top layer of hair from the occipital region of the scalp. Isolate a small thatch of hair (~20 fibres of hair) from underneath this top layer of hair from the occipital region (can use a hair clip to keep the top layer of hair away)
3. Cut the small hair sample off the patient's head as close to the scalp as possible
4. Unfold the piece of aluminium foil and place the cut thatch of hair inside the piece of foil
5. Place a small label with the patient's study identity (ID) over the distal end of the hair thatch (the side furthest away from the scalp)
6. Refold the foil over to completely enclose the thatch of hair
7. Place a study ID label on the folded piece of foil
8. Place the folded piece of foil inside the plastic (e.g. Ziplock®) bag (each Ziplock bag will have a desiccant bag in it and seal the bag.

Given that black children and adolescents naturally have short hair that is curled or kinky, hair is being collected regardless of length. However, for those that are shaved bold, the participant will be enrolled, but study initiation will be deferred in that particular participant for 1 month to allow hair to grow to a length that can be analysed.

Directly administered anti-retroviral therapy

Modified DAART is being administered to randomly selected participants for 3 months (90 consecutive calendar days including weekends and public holidays). These 90 days will be divided into 3 months (months 1, 2 and 3), each month being 30 days. Participants will be visited at home by research assistants who will directly watch them swallow their ARVs and complete a home visit chart, while the participant completes a pill chart. Home visits will be scheduled once a day from Monday to Friday, at the time the participants take their ATV/r. Research assistants will be required to fill in a form when they see the participant swallowing their ARVs and the time they have seen them swallow. Home visits are scheduled as below:

1. Daily for the first 10 days- Monday to Friday

2. Then twice a week for 2 weeks

3. Then once a week for 2 weeks

4. Then fortnightly for the remainder of the 3 months On Saturdays, Sundays and public holidays, each participant will be contacted by cell phone (SMS texts) once a day to be reminded to take their medication. Contacting the participant by cell phone will only be done by the research assistants and recorded on a chart.

   With the exception of ATV/r, all other antiretroviral therapy drugs will be self administered, and the participant will be asked if they took them when the VLW presents for mDAART, or retrospectively for weekends and public holidays. A separate form will be filled in for these drugs.

   Appropriate time for taking ARVs in this study is within 2 hours of the time that the participant is supposed to take his/her medication. Therefore, a dose is recorded as being taken correctly if it was reported taken either within 2 hours before or within 2 hours after the time they are supposed to take that day's particular dose. If more than 2 hours up to <12 hours lapse from the time the participant was supposed to take a dose, that particular dose will be recorded as delayed. If the participant does not take his/her medication for 12 or more hours, that particular dose will be reported as missed.

   If a participant is not found at home, then a direct phone call is made to the participant or caregiver. On the home visit chart, the research assistant will record that the participant was absent. However, the research assistant will inquire if the participant has taken their medication and record appropriately.

   At the end of 90 days allocated to each participant, adherence will be assessed again by measuring hair atazanavir concentrations and self reported adherence questionnaire.

   Research assistants

   The research assistants were recruited from a community based organisation called Child Protection Society (CPS). These research assistants are known as voluntary lay workers (VLWs), and have been trained in research and ethics. They have also been involved in research several times. The VLWs were trained on the protocol by the PI. A VLW assignment form will be filled in when allocating a VLW to a participant. The VLW has to reside in the same community as the participant.

   Posts were advertised openly within CPS and other relevant community-based organisations. Eligibility criteria included:

   1. Prior research experience
   2. Valid training in research
   3. Resident within Harare hospital catchment area
   4. Demonstration of ability to read and write
   5. Ability to travel on foot and public transport to households
   6. Ability to care and nurture for adolescents
   7. Ability to keep information confidential.

   Eligible research assistants were trained before visits to homes began, then refresher courses were done at 2 months after beginning home visits, then after 3 months.

   CD4 count and viral load testing

   CD4 counts, viral loads and PK profiling are done at baseline and repeated after 3 months of DAART and standard care. Patients who were virologically unsuppressed before DAART are expected to re-suppress. Failure to achieve virological suppression after DAART may mean that other factors are contributing to treatment failure, for instance resistance or suboptimal dosing.

   CD4 counts and viral load samples are processed at a private laboratory at no cost to the patient. However, if the patient has had a CD4 count and viral load measured in the previous 2 months then no repeat CD4 and/or viral load will be done. Results of the tests done within the past 2 months will be used for this study.

   Samples for CD4 counts and viral loads are collected in purple top tubes. At least 10mls will be collected from each participant, and gently inverted 8-10 times to prevent clotting. Whole blood in purple top tubes will be stored and transported at temperature between 15 and 30 degrees celcius. Results are available to the participants and their health care givers in real time.

   Viral loads are measured yearly or when indicated for free to patients registered at Harare hospital opportunistic clinic (OI) clinic. This has become standard practise in the country at institutions where a viral load machine is available. CD4 counts are measured twice per year for each patient who has been on ART for a year or more at a cost.

   Resistance testing and genotyping

   Participants who remain unsuppressed despite DAART and adherence counselling proceed to have resistance testing done. A lot of studies have shown that protease inhibitors have high genetic barrier to clinically significant mutations. However, this does not rule out the presence of clinically significant PI mutations. Two 4ml whole blood samples are collected in purple top tubes, and gently inverted 8-10 times to prevent clotting.

   Keeping track of samples

   A log chart will be available. Each time a sample is taken (hair or blood), the sample is logged onto the log chart. The date and time the sample leaves the OI clinic will be recorded. The date and time the laboratory receives the sample will also be recorded on the same log chart.

   Measurement of adherence in participants

   Self-reported adherence will be measured using AIDS Clinical Trials Group (ACTG) adherence follow-up questionnaire (QLO702) and visual analogue scale (VAS). These tools were validated, and found to have high sensitivity and specificity in clinical trials.

   Defining virological and immunological treatment failure

   Treatment failure will be defined according to modified World Health Organization (WHO) 2012 criteria:

   Virological failure

   • <1.0 log10 decrease in HIV RNA copy number from baseline after 8-12 weeks of therapy, or

   • HIV RNA ≥1000 copies/mL after 6 months of therapy Immunological failure Fall of CD4 count to baseline (or below) OR

     50% fall from on-treatment peak value OR

   Persistent CD4 levels below 100 cells/mm3 (Without concomitant infection to cause transient CD4 cell decrease).

   Please note- All blood samples are labelled with the participants 0I number, study number and the day the sample was collected. The use of these identity numbers is meant to keep the participants' identity hidden from the study and laboratory personnel while preventing sample mix up. All samples will be attached to a laboratory request form.

   Study visit schedule

   Visit 0 Visit 1

   90 days of DAART

   (intervention arm)

   90 days of standard care (SC)

   (control arm) Visit 2 Visit 3

   Eligibility

   screening Consent and assent

   Enrollment

   Randomization

   Adherence questionnaire

   Participant characteristics questionnaire

   Hair sample collected Adherence questionnaire

   Hair sample collected Genotyping for eligible participants

   Study setting

   Harare central hospital is arguably the largest referral hospital in Zimbabwe. Because congestion could occur, patients that are doing well on first line treatment are decentralised to municipal clinics close to where the patients reside. This move is not only meant to decongest the hospital, but also to shorten the distance travelled by patients to seek health care, cut on transport costs and remove technical barriers in accessing health care. However, patients that remain unwell are on fluconazole maintenance treatment and all those on second line treatment continue to be monitored at Harare hospital OI clinic. If decentralised patients develop treatment failure or fall ill while being seen at municipal clinics, they are referred back to Harare hospital OI clinic for specialist care.

   Harare hospital OI clinic is situated within the paediatric hospital in Harare hospital. It is open to patients from Mondays to Thursdays, excluding public holidays. The clinic caters for children that are HIV positive up to 18 years of age. Once they turn 19, they are transferred to the adult opportunistic infections clinic at the same hospital. On each visit to the clinic, patients that were commenced onto ARVs or cotrimoxazole prophylaxis receive a maximum of 2 months supply. This means that each patient enrolled in the clinic has to return to the clinic at least once in 2 months for drug refill. However, if any patient is taken ill or has any other concern, they are free to visit the clinic to be seen by the doctors, nurses or counsellors outside their review date.

   On presenting to the clinic, each patient is screened by nurses whether they should be seen by the doctors or just be given a prescription so they can collect medication. Patients requiring counselling are referred to counsellors in the clinic. All patients eventually queue up at the pharmacy in the OI clinic with their hand held and hospital records for drug supply.

   Data collection and storage

   Information about patient is collected from 3 sources: the participant and/or guardian, patients' hand held record and patients' hospital record. This move is meant to ensure that information gathered is as accurate as possible.

   Information is collected using interviewer based questionnaires and kept in accessible files. The amount of time spent on each interview is minimized as much as possible. This is to enable many participants to be interviewed and to avoid delaying them unnecessarily. Each participant will have a file (hard copies) with all his/her information, including printed questionnaires with their information, kept in a fireproof, lockable cupboard in the department of clinical pharmacology, in the college of health sciences of the University of Zimbabwe. Upon completion of the study, data will be kept for at least 3 years. Only the principal investigator and co-investigators will have access to soft and hard copies of data.

   Data will be entered into research electronic data capture (REDCap) software, for easier analysis of data.

   Loss to follow-up (LTFU)

   A participant is considered lost to follow-up if:

   1. A participant misses 2 consecutive scheduled hospital visits
   2. The participant cannot be found at the addresses they provided on the participant locator form for 30 consecutive days
   3. The participant cannot be reached on the phone numbers provided on the participant locator form for 30 consecutive days
   4. If at the addresses provided no one knows where the participant relocated to or the participant cannot be contacted at the new location

   When a participant decides to discontinue participating, or when a guardian decides to discontinue their child's participation from the study, that case is recorded as a withdrawal. In such a case, a withdrawal form will be filled in and filed in the participant's folder.

   Participants lost to follow-up will be replaced. A lost to follow-up form will be filled in and placed into the participant's file.

Inclusion Criteria:

1. Parents/guardian willing to consent
2. Child willing to provide assent
3. Documented HIV positive antibody or antigen test
4. Child knows their HIV status
5. Aged between 6 and 18 years (that is, from the day of their 6th birthday up to the eve of their 18th birthday)
6. Registered at Harare hospital paediatric opportunistic infections clinic
7. On second line treatment (ATV/r based)
8. Have taken the above named second line treatment for at least 6 complete, consecutive months
9. Has virological and immunological treatment failure as defined by WHO 2012 criteria

Exclusion Criteria:

1. Patients registered at other health centres who have been referred for specialist care at Harare hospital paediatric opportunistic infections clinic
2. On ATV/r as first line treatment
3. Patients who do not want to be followed up at home.
4. On anti-tuberculosis (TB) treatment