Study Design
The study is a randomized, double-blind, placebo-controlled clinical trial of oral zinc supplementation (10 mg zinc sulfate) or placebo, administered once daily, to reduce the incidence of severe or invasive infection in Ugandan children with confirmed sickle cell anemia (SCA) between 1 and less than 5 years of age. The participant flow diagram is shown in Figure 1.
Study Objectives
The primary objective of the study is to determine if zinc supplementation reduces severe or invasive infections in Ugandan children 1.00-4.99 years of age with SCA. The study will also assess four secondary outcomes that may be affected by zinc supplementation in children: 1) incidence of all clinically diagnosed infections; 2) incidence of culture or PCR-confirmed bacterial infections, 3) incidence of vaso-occlusive crises (VOC), and 4) change in height-for-age z-score.
Eligibility Criteria
Inclusion Criteria
Inclusion criteria include the following:
- Documented sickle cell anemia (HbSS or HbS/b0 thalassemia supported by hemoglobin electrophoresis)
- Age range of 1.00-4.99 years, inclusive, at the time of enrollment
- Weight at least 5.0 kg at the time of enrollment
- Willingness to comply with all study-related treatments, evaluations, and follow-up
Exclusion Criteria
Exclusion criteria include the following:
- Known other chronic medical condition (e.g., HIV, malignancy, active clinical tuberculosis)
- Severe malnutrition determined by impaired growth parameters as defined by WHO (weight for length/height or height for age z-score <-3, using WHO growth standards)
Study Setting
Study participants will be recruited from the Nalufenya Sickle Cell Clinic (NSCC) in the Children’s Ward at the Jinja Regional Referral Hospital in Jinja, Uganda. NSCC serves a region of high malaria transmission intensity along the shores of Lake Victoria. The clinic has over 3500 registered SCA children and is run by two pediatricians, assisted by medical officers, nurses and counselors. Nalufenya Children’s Ward, part of Jinja Regional Referral Hospital, has been the site of several epidemiologic and clinical studies of children [53].
Treatment Groups
Children will be randomly assigned to receive dispersible zinc sulfate tablet (10 mg) or identical placebo tablet in a 1:1 ratio. Zinc tablets will be certified as manufactured under good manufacturing process (GMP). Each tablet is designed to dissolve completely in a small amount of water or breast milk in <3 minutes, removing the need for small children to swallow a pill. Caregivers will be instructed to place the pill in 5-10 mL of clean water or breastmilk, stir gently, and wait three minutes until the pill is completely dissolved. Both active tablets and placebo will have the same flavoring added to enhance palatability
Randomization and Blinding
Block randomization will be used for this study. Children will be randomized (in blocks of 8) into treatment groups by order of entry in the study, based on a pre-determined blinded randomization list created and managed by an Indiana University study data manager. Treatment group will be provided to the study pharmacist, who will know whether the child is randomized to group A or B, but will not know which group is zinc or placebo. The study pharmacist will have identical appearing A or B tablets and will provide the appropriate medication to the child. The designation of A or B on packets will be removed by the pharmacist prior to dispending the study drug so the medication will appear identical to study staff and parents/children. The child’s study identification number will be recorded and treatment group may only be determined by comparing the child’s study id to the blinded list, which only the Indiana University data manager will have access to until the study is completed or stopping rules are reached.
Outcome Measures
The primary outcome to be assessed in the ZIPS trial is the incidence of severe or invasive infections. At all unscheduled sick visits children will be evaluated for clinical evidence of infection by taking clinical history and exam, and diagnostic work up. Children with a measured axillary temperature of ≥37.5C will have blood obtained for a malaria smear and a blood culture for a measured fever of ≥38C. Children with history of fever or temperature ≥37.5˚C and age-specific tachypnea and cough will have a chest radiograph obtained.
Severe or invasive infections will include: abscesses, bacteremia, cellulitis, diarrhea, dysentery, malaria, meningitis/encephalitis, osteomyelitis, pharyngitis/tonsillitis, pneumonia/acute chest syndrome, sepsis, acute sinusitis using standard definitions (Table 1). Other common infections in this age group (e.g., acute upper respiratory infection (URI), otitis media, conjunctivitis, tinea capitis, tinea corporis) will be recorded, and included (along with the severe and invasive infections) in the category of “clinical infections”. Viral infections with a well-defined clinical picture (e.g., measles, varicella) will be defined by clinical signs and symptoms.
Secondary outcomes of the study include: 1) Clinical infection, as described above: 2) Culture confirmed bacterial infection (e.g., bacteremia, urinary tract infection, tonsillitis, abscess, osteomyelitis, meningitis) or PCR-confirmed infection with Chlamydophila pneumoniae or Mycoplasma pneumoniae (from nasopharyngeal swab) in children with pneumonia/acute chest syndrome. Pneumonia (acute lower respiratory infection, ALRI) will be defined as: history of fever or measured axillary temperature ≥37.5˚C, with age-specific tachypnea, cough, and an infiltrate and/or effusion on chest x-ray consistent with pneumonia. 3) Vaso-occlusive crises (VOC): pain with the requirement for oral morphine or IM diclofenac, per Sickle Cell Clinic (SCC) guidelines. 4) Change in height for age z-score, from enrollment to 12 months follow up, calculated using WHO standards.
Safety
Zinc is approved for use for the treatment of diarrhea, where it has been shown to reduce the duration of diarrheal illness [54]. It is widely available in Uganda and has a well-established safety record. The primary side effect of zinc is vomiting. For children who do have problems with vomiting, parents will be told to give the zinc with food, as this can decrease vomiting. Zinc may interfere copper absorption [55]. We will measure copper levels from samples at baseline and 12 months, to see if copper levels are affected by one-year of zinc supplementation.
Adverse events will occur commonly in a trial involving children with SCA, although the majority of events are likely due to the underlying disease process and risk of infections in childhood and not to study medication. Adverse events (AEs) will be defined according to Good Clinical Practice (GCP) and will be logged prospectively and tabulated at the end of the study, disaggregated by study arm. AEs will be defined using the Common Terminology Criteria for Adverse Events (CTCAE) Version 4.0 available since 2009 (Table 2 and 3), where AEs are categorized by organ system and graded by severity. Serious adverse events (SAEs) will adhere to standard definitions (any life-threatening event hospitalization or death); however, since hospitalization is common in children with SCA, we will use hospital stay of more than 7 days to define hospitalization-related SAE in this study. This is based on the knowledge that children who are admitted with sickle cell-related conditions (such as anemia requiring transfusion, acute chest syndrome, and stroke) have an average length of stay of approximately 7 days. A SAE for this study will therefore include hospitalization for more than 7 days.
In addition to the ethical oversight provided by both Ugandan and North American institutions, an external and independent Data and Safety Monitoring Board (DSMB) has been convened to oversee the trial. The study may be discontinued at any time if the DSMB or study team feels that it is in the best interests of study participants to do so. Stopping rules for SAEs will be developed by the study biostatistician in conjunction with the DSMB. Stopping rules will be created only for SAEs, not efficacy or futility, for this study because there is no perceived benefit to an interim analysis of efficacy/futility in this small cohort with a short treatment duration.
Sample Size and Power Calculation
We will assess reduction in incidence of severe or invasive infections, with or without culture or PCR confirmation. Sample size is based on incidence of severe or invasive infections, using a baseline rate of 0.71 infections/child/year, derived from data of children from the NOHARM study [5]. Power calculations assume an alpha of 0.025 for a one-sided test or 0.05 for a two-sided test. With an incidence of 0.71 severe or invasive infections/year in the placebo group, a sample size of 250 children (with a 10% loss to follow-up) will have 80% power to detect a decrease of ≥40% in severe or invasive infection incidence over the 12-month study period. This decrease is smaller than the 47-88% reduction in clinical infection incidence in adolescents and adults in previous studies [40-42], so the study sample size should allow us to detect the expected effects of zinc on infection incidence.
Proposed Analysis
We hypothesize that the incidence of infection in the zinc supplemented group will be ≥40% lower than that in the placebo group. Incidence of clinical infection will be compared using Poisson or negative binomial regression analysis. Other factors that are potentially associated with the risk of infection will be included in the regression models as covariates. Similar analyses will be conducted for incidence of culture or PCR-confirmed bacterial infections, malaria, VOC and other sickle-related clinical complications, SAE and AE. Frequency of infection, culture or PCR-confirmed bacterial infection, VOC, AE and SAE will also be compared using c2 analyses.
Study Assessments
At enrollment, children will have a standard history and physical exam performed. At baseline and 12- month visits, children will receive a CBC with differential, reticulocyte count, hemoglobin electrophoresis, and plasma stored for zinc testing. Socio-economic status and dietary history will be assessed using a dietary instrument developed and validated for use in a Ugandan population. Urine samples will be collected in zinc-free containers at enrollment and 12-month follow-up in 100 randomly selected children to test for urine zinc levels, to assess degree of urinary zinc loss in study children at baseline and after 12 months of zinc or placebo treatment. Stool samples will also be collected from children who are able to provide them for future microbiome testing.
Children in the study will have scheduled clinic visits at 1, 3, 6, 9 and 12 months to assess study adherence by pill counts, evaluate adverse events, assess weight and height, and to refill zinc or placebo tablet supply. Parents or guardians will be asked to bring their children to the NSCC for any illness. Children with illness will be evaluated clinically and be managed according to National and local SCC treatment protocols. Standardized definitions will be provided for the most common infections (Table 1). Children with pneumonia/acute lower respiratory tract infection (ALRI) will have a nasopharyngeal (NP) swab collected for later testing by PCR for Mycoplasma pneumoniae, Chlamydophila pneumoniae, Bordetella pertussis and 17 viral upper respiratory infection (URI) pathogens. Children who have clinical signs of pneumonia will receive a chest x-ray. Other labs will be collected as needed for clinical diagnosis, and a blood sample collected and stored for future infectious pathogen or inflammatory response testing. A summary of the schedule of enrollment, interventions, and study assessments is shown in Figure 2.
Laboratory testing
Plasma zinc and copper levels and urine zinc levels will be tested in the baseline and 12-month samples at the Wright Lab at Mount Sinai Hospital, New York, NY, or a similarly certified lab for zinc testing. Blood and urine samples will be collected using a tight trace metal specific sample collection protocol to minimize contamination. NP swab specimens will be analyzed using the FilmArrayÒ Respiratory Panel (RP; BioFire Diagnostics, Salt Lake City, UT) to test for C. pneumonaie, M. pneumoniae, B pertussis and 17 viral URI pathogens. Cultures will be done in Bactech bottles (Becton Dickinson, Sparks, MD), Microscopy for Plasmodium species by thick and thin smear, with parasite quantification, will be performed as previously described [56].