Early Life Interventions for Childhood Growth and Development in Tanzania (ELICIT): a protocol for a randomised factorial, double-blind, placebo-controlled trial of azithromycin, nitazoxanide and nicotinamide

Introduction In many developing areas in the world, a high burden of enteric pathogens in early childhood are associated with growth deficits. The tryptophan-kynurenine-niacin pathway has been linked to enteric inflammatory responses to intestinal infections. However, it is not known in these settings whether scheduled antimicrobial intervention to reduce subclinical enteric pathogen carriage or repletion of the tryptophan-kynurenine-niacin pathway improves linear growth and development. Methods and analysis We are conducting a randomised, placebo-controlled, factorial intervention trial in the rural setting of Haydom, Tanzania. We are recruiting 1188 children within the first 14 days of life, who will be randomised in a 2×2 factorial design to administration of antimicrobials (azithromycin and nitazoxanide, randomised together) and nicotinamide. The nicotinamide is administered as a daily oral dose, which for breast-feeding children aged 0–6 months is given to the mother and for children aged 6–18 months is given to the child directly. Azithromycin is given to the child as a single oral dose at months 6, 9, 12 and 15; nitazoxanide is given as a 3-day course at months 12 and 15. Mother/child pairs are followed via monthly in-home visits. The primary outcome is the child’s length-for-age Z-score at 18 months. Secondary outcomes for the child include additional anthropometry measures; stool pathogen burden and bacterial microbiome; systemic and enteric inflammation; blood metabolomics, growth factors, inflammation and nutrition; hydrogen breath assessment to estimate small-intestinal bacterial overgrowth and assessment of cognitive development. Secondary outcomes for the mother include breastmilk content of nicotinamide, other vitamins and amino acids; blood measures of tryptophan-kynurenine-niacin pathway and stool pathogens. Ethics and dissemination This trial has been approved by the Tanzanian National Institute for Medical Research, the Tanzanian FDA and the University of Virginia IRB. Findings will be presented at national and international conferences and published in peer-review journals. Protocol version 5.0, 4 December 2017. Protocol sponsor Haydom Lutheran Hospital, Haydom, Manyara, Tanzania. Trial registration number NCT03268902; Pre-results.

Introduction: In many developing areas in the world, a high burden of enteric pathogens in early childhood are associated with growth deficits. The tryptophan-kynurenine-niacin pathway has been linked to the enteric inflammatory response to intestinal infections. However, it is not known in these settings whether scheduled antimicrobial intervention to reduce subclinical enteric pathogen carriage or repletion of the tryptophan-kynurenine-niacin pathway improves linear growth and development.
Methods and analysis: We are conducting a randomized, placebo-controlled, factorial intervention trial in the rural setting of Haydom, Tanzania. We are recruiting 1188 children within the first 14 days of life, who will be randomized in a 2 x 2 factorial design to administration of antimicrobials (azithromycin and nitazoxanide, randomized together) and nicotinamide. The nicotinamide is administered as a daily oral dose, which for breast-feeding children aged 0-6 months is given to the mother and for children aged 6-18 months is given to the child directly.
Azithromycin is given to the child as a single oral dose at months 6, 9, 12, and 15; nitazoxanide is given as a 3-day course at months 12 and 15. Mother/child pairs are followed via monthly inhome visits. The primary outcome is the child's length-for-age Z-score at 18 months. Secondary outcomes for the child include additional anthropometry measures; stool pathogen burden and bacterial microbiome; systemic and enteric inflammation; blood metabolomics, growth factors, inflammation and nutrition; hydrogen breath assessment to estimate smallintestinal bacterial overgrowth; and assessment of cognitive development. Secondary outcomes for the mother include breast milk content of nicotinamide, other vitamins and amino acids; blood measures of tryptophan-kynurenine-niacin pathway; and stool pathogens.  • This study assesses the efficacy of two different intervention approaches targeting early childhood growth and development in a rural site in sub-Saharan Africa with a high degree of both enteric pathogen carriage and stunting.
• This study assesses intervention with nicotinamide, targeting the tryptophan-kynurenineniacin pathway, which may have important roles in intestinal bacterial content and immune and metabolic response.
• In addition to the primary outcome of length-for-age Z-score, this study measures extensive secondary outcomes related to infection, nutrition, metabolism and cognitive development.
• Due to sample size constraints, one limitation of the study is that two different antimicrobial interventions (azithromycin and nitazoxanide) are randomized together, limiting the ability to determine any individual contribution of either of these to study outcomes. worldwide, including the region around Haydom, Tanzania. 3 4 Haydom is in a rural, semi-arid area with a high degree of poverty. Children in Haydom followed in MAL-ED experienced a 70% rate of stunting by 18 months of age. 5 Data from MAL-ED also showed that linear growth deficits correlated with lower scores for cognitive development, 6 potentially through shared barriers to optimal growth and brain development. 1 7

Antimicrobials to improve childhood growth
In MAL-ED, children in the Haydom site had a high burden of infection with intestinal pathogens starting early in life. High carriage of enteropathogens, even in the absence of diarrhea, was associated with a 1.36 increased odds of having lower length for age. 3 As for the specific enteropathogens, among the most prevalent pathogens at age 6 months were enteroaggregative E.coli (48%) and Campylobacter (27%), 8 each of which were associated with a 0.85 cm and 0.83 cm, respectively, shortfall in height attainment. 9 Both of these infections are treatable with azithromycin. Other highly prevalent pathogens by age 6 months in the MAL-ED study included Giardia (22%) and Cryptosporidium (6%), and early Giardia infection was also associated with growth shortfalls. 10 Both of these protozoal infections are treatable with nitazoxanide. These observations raise questions as to whether such antibiotics may be useful in ameliorating growth deficits.
A meta-analysis of 10 randomized controlled trials showed that antibiotic use increased height by 0.04 cm/month. 11 Another study demonstrated that randomized administration of azithromycin to asymptomatic young children in Ethiopia was associated with a halving of mortality, 12 suggesting overall effects of suppressing bacterial pathogens in areas with high rates of infectious disease. However results from individual studies of azithromycin mass drugadministration on growth have not shown consistent results. 13 Among children with cryptosporidial infection in Zambia, treatment with nitazoxanide resulted in reduced mortality, while linear growth was not assessed. 14 Therefore, the role of these antimicrobials for linear growth remains unclear.

Nicotinamide to improve childhood growth
Another potential means of improving growth is by targeting the tryptophan-kynurenineniacin pathway ( Figure 1). In this pathway, dietary tryptophan is absorbed and then metabolized by indoleamine 2,3-dioxygenase (IDO) to kynurenine and later to nicotinic acid, a key precursor to NAD+. The importance of this pathway has been supported by recent work identifying mutations in the NAD+ pathway as a cause of birth defects in humans; in animal models with these mutations, birth defects were prevented via supplementation with niacin. 15 Deficiencies in this pathway have been noted in areas where diets are heavily maize-based, leading to pellagra. Infants at the Haydom site in MAL-ED were noted to have inadequate dietary intake of multiple vitamins 16 ; while deficiency in niacin was not assessed, individuals in the region around Haydom eat a diet based heavily on maize, with potentially low bioavailability of tryptophan.
Basic science models have demonstrated that tryptophan deficiency is associated with reduced growth velocity. 17 Abnormalities in the tryptophan-kynurenine-niacin pathway are also associated with intestinal inflammation, including during response to intestinal pathogens. 18 19 In animal models, an inability to transport tryptophan into enterocytes results in diarrhea, while provision of nicotinamide (a form of niacin) partly restores function of mTOR in enterocytes, increasing secretion of antimicrobial peptides and reducing diarrhea, 18 19 suggesting effects on diarrheal pathogens. IDO activity is increased during inflammation, including in inflammatory bowel disease 20 21 and during infections; 22 this appears to be a maladaptive in most cases, as kynurenine has potential immunosuppressive effects. 23 24 In MAL-ED, children at the Haydom site with higher ratios of kynurenine:tryptophan were noted to have poorer linear growth. 24 Given that 1) children in the Haydom area have a maize-based diet low in tryptophan, a key precursor to niacin/nicotinamide and NAD+, 2) preclinical studies demonstrate improvements in antimicrobial peptides and reduced diarrhea with nicotinamide supplementation, 3) an increase in nicotinamide as a precursor of NAD+ should cause reduced need for the tryptophankynurenine-niacin pathway, reducing activity of IDO and thereby reducing kynurenine, and 4) higher kynurenine:tryptophan is associated with poor growth, we reasoned that intervention with nicotinamide among children in this region would improve linear growth.

Study hypotheses
We hypothesized that among children with a high degree of stunting and enteric pathogen burden in this region, intervention with 1) antimicrobials and/or 2) nicotinamide would increase linear growth. We further set out to interrogate the potential mechanisms behind these effects. In the case of antimicrobial intervention, we hypothesize that intervention with azithromycin and nitazoxanide (relative to placebo) would result in a decrease in enteropathogen burden that would be associated with an increase in growth. In the case of nicotinamide (relative to placebo), we hypothesize that intervention will result in a reduction in the kynurenine:tryptophan ratio, an increase in availability of nicotinamide for NAD+ synthesis and associated increases in growth.

OBJECTIVE
The primary objectives of this randomized factorial, double-blind, placebo-controlled trial are to determine if interventions with 1) antimicrobials and 2) nicotinamide are associated with F o r p e e r r e v i e w o n l y 8 increased length-for-age z-score (LAZ, relative to placebo-treated children) by age 18 months.
Specifically, the antimicrobial intervention consists of azithromycin administered as a single dose at age 6,9,12, and 15 months and nitazoxanide as a 3-day course at age 12 and 15 months (with both antimicrobials randomized together). The nicotinamide intervention consists of nicotinamide 250 mg daily administered daily by mouth to the child's mother during breast feeding 0-6 months (and thus passed on to the child via breast milk 25

Eligibility
Participants are eligible if the mother is >18 years and the child is age <14 days and was born in the Haydom catchment area (defined for study purposes as 25 km radius of Haydom Lutheran Hospital).
Exclusion criteria are as follows: maternal inability to adhere to the protocol, multiple gestation, significant birth defect or neonatal illness, weight <1,500 g at enrollment (age < 14 days), lack of intent to breastfeed infant and plan to move from area within the next 18 months. Recruitment occurs via survey by field teams of surrounding villages and via input from local community healthcare workers. Pregnant women are identified and informed regarding the study. Pregnant women in the third trimester may provide informed consent for the study but are not enrolled until the initial assessment visit of the infant to determine eligibility (which includes ascertainment of infant weight). If informed consent is not obtained during pregnancy, it is obtained at enrollment.

Enrollment
Infant/mother pairs are enrolled during the initial study visit to their homes between birth and day of life 14, after eligibility criteria are confirmed, including measurement of infant weight which should be 1.5 or above. Informed consent is provided at this time, if it was not obtained during pregnancy. Eligibility (besides current weight) is confirmed and the mother/child pair is randomized in the study.

Randomization
Randomization occurs at the time of enrollment. Block randomization was used in groups of eight for the four intervention combinations (antimicrobial + nicotinamide, antimicrobial alone, nicotinamide alone, all placebo). Study teams assign participant study number prior to the initial study visit.

Intervention
As a factorial study design, the two intervention domains (antimicrobial and nicotinamide) are randomized separately. The antimicrobial intervention consists of two medications that are randomized together: azithromycin and nitazoxanide (vs. placebos of both). Azithromycin 20 mg/kg (vs. placebo) (both manufactured by Universal Corp, Kenya) is Prior studies have demonstrated that nicotinamide is present in breast milk in quantities that are associated with serum levels. 25 Haydom-specific data from MAL-ED demonstrated a high prevalence (96%) of breast feeding until children were 6 months old. 5 From months 6-18 the child receives sachets of nicotinamide 100 mg by mouth daily.
Mothers are instructed to mix the nicotinamide powder with a small volume of age-appropriate food (e.g. infant cereal) to give to the child.
For both the maternal nicotinamide pills and the children's nicotinamide sachets, doses are given as a 60-day supply every two months. When a new supply is provided, the family is asked for the remainder of doses not consumed the prior 60 days. These unused doses are counted and recorded by pharmacy staff to determine adherence.

Questionnaires
Study staff use standardized questionnaires designed for the study, in many cases adapted from questionnaires used in the MAL-ED study. These include questions asked at each monthly visit regarding breastfeeding, use of complementary foods, childhood illness and treatment (including whether an antibiotic was used and if so which one), types of food sources given, and food insecurity. Additional questionnaires are asked at other specific time periods, including aspects of the child's home environment (completed at the first month's visit) and the vaccines that the child has received (completed at the 12-month visit).

Anthropometry measures
All field team members are trained in the accurate measurement of anthropometry indices. The child's weight is measured at the initial visit and at each monthly visit through age 18 months. The child's length and head circumference are measured at the initial visit and every three visits thereafter through 18 month. The mother's weight and height are measured at month 10-11. All weights are measured using a digital scale that undergoes quality assessments weekly. Infants are weighed on an infant tray with the infant nude or wearing a dry diaper or standing, when able. Lengths are measured by two field team members with the child lying flat on a measuring board with a fixed, perpendicular board against which the head is placed and a second adjustable perpendicular board is placed against the feet. Two measurements are obtained; if these are within 2 mm, the average of the two is used; if these are not within 2 mm, a third measurement is obtained and the average of the two closest measurements is used. Head circumference is measured by placing a non-distensible tape around the child's head just above the eyebrows, above the ears, and around the biggest part of the back of the head. Maternal weight is measured using a standing scale. Maternal height is measured using a portable stadiometer. For quality assurance, 5% of infant weights, lengths, and head circumferences are measured by a supervisor skilled in measurement to compare to the field team's measurement.

Developmental assessments
To assess maternal-child interaction, an important predictor of early childhood development and health, 26 27 The Observation of Maternal Child Interaction (OMCI) 28 F o r p e e r r e v i e w o n l y 12 assessment is collected at 12 months of age. This observation includes a 5 minute observation of mother and child interacting over a picture book, with ratings given for positive affect, verbal statements, sensitivity, scaffolding, language stimulation, focus and mutual enjoyment, among others.
At 18 months, developmental assessment is performed by trained assessors using two standardized scores, the Malawi Developmental Assessment Tool (MDAT) 29

and the Bayley
Scales of Infant and Toddler Development, cognitive assessment (Bayley). 30 The MDAT is a validated score that tests multiple childhood development domains, including gross motor, fine motor, language and social development that has been validated in this age range in Malawi and intended for use in rural sub-Saharan settings. Briefly, this assessment was designed to be a culturally-relevant method for examining child development across multiple domains using objects and tools typically available in rural or lower-resource settings. Both direct assessment and caregiver report of skills go into the score. The Bayley cognitive assessment involves a series of tasks presented to the infant, which assess for play, reasoning, and mental representation. This is a commonly used assessment for early childhood and has been used in many research sites around the world. 6 31

Specimen collection
Breast milk (months 1 and 5): Participating mothers express a maximal volume of breast milk from one breast (up to 40 mL), using hand expression under low-light conditions. Breast milk will be collected into foil-covered sterile containers and transported on ice to the central laboratory center within 8 hours for archiving of whole milk at -80 degrees.
Blood (months 12 and 18; in a subset months 2 and 8): Blood is drawn from an experienced phlebotomy team using a butterfly needle and syringe and placed in micro-containers for transport to the laboratory for processing into serum and plasma and storage at -80 degrees. In Stool (months 6, 9, 12, 15, and 18): Stool is collected in diapers that are provided just prior to the stool collection date. Mothers or caregivers are instructed to collect approximately 3-4 spoonfuls of stool and place it within 1 hour into a labeled stool container, seal in a plastic bag, and place inside a transport box with cold packs. The stool collection will be timed prior to a planned study visit, such that the field-worker can collect the sample and the laboratory process the specimen shortly thereafter. Stool samples are then stored at -80 degrees. A subset of specimens, prior to freezing will be cultured for E.coli and screened for antibiotic resistance (AMR).
Urine (months 6, 12, and 18): Urine is collected by placing a bag over the child's genitalia at the beginning of a study visit. The child is given adequate hydration, via breast feeding if possible.
Collected urine is then transferred in a sterile container that is stored on ice prior to transport to the research site and then stored at -80 degrees.
Saliva (maternal and infant at month 12): Maternal and child saliva is collected into a vial containing storage reagent and stored at -80 degrees for genomic assessment.
Breath (subset of participants at months 6, 9 and 12): Glucose hydrogen breath testing is performed in a subset of children as an assessment of small intestine bacterial overgrowth (SIBO), 32 which has been associated with intestinal inflammation and linear growth shortfall. 33 Children are fasted for 2 hours prior to the initiation of the breath test. They are then given 1 g/kg actual body weight of a glucose dissolved in a 1 g/5ml solution. Following ingestion of the

Laboratory procedures and storage
Upon receipt of the samples listed above, the laboratory team at the research site prepares aliquots of samples for storage at -80 degrees for local testing or shipping to additional testing facilities as shown below.

Rationale
During the first 6 months of participation, the only means of the child receiving intervention is through maternal delivery of nicotinamide in breast milk. However, nicotinamide in vivo is both derived from tryptophan and converted to multiple other vitamers. 34 As such, we will assess content of nicotinamide, vitamers and tryptophan in breast milk at two time points during the intervention.
Maternal breast milk samples collected at months 1 and 5 will be tested at the United States Department of Agriculture laboratory at the University of California -Davis. Planned LC-MS analyses includes nicotinamide, thiamin, riboflavin, flavin adenine dinucleotide, pyridoxal, pyridoxine, biotin and pantothenic acid concentrations, and on a random subset (n=600 samples) niacin vitamer, and free and total tryptophan concentrations. Blood testing will include assessment of multiple measures both to verify safety of the interventions and interrogate the effects of the intervention on physiological processes related to linear growth. Safety assessment includes testing of general tests of health with basic metabolic panel, liver function and enzyme testing, and complete blood count. This will assess for differences between intervention groups in measures of kidney, liver and hematopoetic health. Additional tests to interrogate the physiologic effects of the intervention include assessment of metabolic effects (using metabolomics) and assessment of levels of factor related to growth (insulin-like growth factor-1 [IGF-1] and collagen X).

Blood
Complete blood count (CBC): To assess general safety of the nicotinamide dosing in the first 100 infants at months 2 and 8. CBC is performed at Haydom Lutheran Hospital clinical on a Swelab Alfa hematology analyzer (Boule Diagnositics, Stagna, Sweden; this machine has quality control assessed daily).
Basic metabolic panel (BMP), aminotransferases and bilirubin: To assess general safety of the nicotinamide dosing in the first 100 infants at months 2 and 8, additional laboratory measures will be performed on serum samples at the Kilimanjaro Christian Medical Center. BMP, bilirubin, alanine aminotransferase (ALT) and aspartate aminotransferase (AST) will be assessed using a Cobas Integra 400 Plus biochemistry analyzer (Roche Diagnostics, Germany), according to the manufacturer's instructions. 35 Insulin-like growth factor-1 (IGF-1) and high-sensitivity C-reactive protein (hsCRP): To interrogate these interventions for changes in growth factor and low-grade systemic inflammation, respectively, IGF-1 and hsCRP will be tested at 12   16 Collagen X: We will assess for intervention-associated changes in collagen X (a biomarker of rapid skeletal growth) using an ELISA developed at Shriners Hospital for Children in Portland, OR, and Oregon Health and Science University. 36 Metabolomics: The tryptophan-kynurenine-niacin pathway will be assessed at 12 and 18 months via metabolomics analyses of serum and plasma using 1 H nuclear magnetic resonance (NMR) spectroscopy and ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) at Imperial College, London. Assessments include the following: 1) metabolites of the tryptophan-kynurenine-niacin pathway to assess for (among other items) changes in the tryptophan:kynurenine ratio in the setting of nicotinamide treatment; 24 2) plasma bile salts, which have implications for risk of infection; 37 3) untargeted metabolomics analysis, providing information on the global metabolic status of the individual including pathways related to the metabolic activity of the gut microbiota and their biochemical interactions with the host; 38 and untargeted UPLC-MS based lipidomics to assess for differences in lipid metabolism by nutritional status.

Stool
Stool testing will include assessment of the following: Microbiome: Stool bacterial content will be assessed at in a subset of 200 participants at the University of Maryland. Total stool DNA will be isolated by bead beating method. A library will be created each specimen. The libraries will be multiplexed and sequenced by Illumina HiSeq.
The resulting reads will be cleaned, binned by specimen ID, separated in operational taxonomic units (OTUs) by DNAclust 39 and compared using metagenomeseq. Metagenomeseq normalizes the data and introduces a correction factor to account for under-sampling of OTUs. The normalized data will be compared using odds ratios and multiple t-tests, with p-values adjusted for multiple testing.
Taqman Array Cards (TAC): Pathogen burden will be assessed using TAC cards ( Figure 2) as performed previously. 40 41 42 This will be performed at in the laboratory at Haydom Global Health Research Center using. Briefly, stool samples undergo nucleic acid extraction via the QIAamp Fast Stool DNA Mini kit (Valencia, CA). Nucleic acids are mixed with AgPath-ID One-Step RT-PCR buffer (Thermo Fisher) and nuclease free water to a 100ul final volume and run on a custom-designed TAC 40 : Antibiotic resistance: We will perform stool bacterial culture on a subset of participants to monitor for development of antimicrobial resistance, which is a valid concern of any mass antimicrobial administration. Briefly, stool will be cultured on MacConkey agar and 5 lactose fermenting E. coli colonies pooled and tested for AMR using disc diffusion to screen for ESBL, AmpC, carbapenemase and susceptibility to antibiotics commonly in use locally.
Azrythromycin susceptibility will be tested by E-Test. 43 Myeloperoxidase: As an assessment of stool inflammation, myeloperoxidase will be assessed at months 6, 12 and 18. Testing will be performed at the Global Health Research Center laboratory in Haydom using a commercially-available assay (Alpco, Salem, NH).

Urine
Urine samples at 6, 12 and 18 months will assessed alongside serum and plasma samples at Imperial College, London for untargeted metabolomics analysis by NMR spectroscopy and UPLC-MS, as described above. 38 For a more complete assessment of these interventions on intestinal inflammation in this setting, we will test mothers and children for secretor status of fucosyltransferase 2 (FUT2), a regulator of blood group antigens on the intestinal mucosa, as non-secretor status has been associated with infections 44 and other forms of colitis. 45 DNA for mothers and children will be extracted from salivary samples at UVa and FUT2 genotype will be determined by dot blot assays in a hierarchical method based largely on Lewis Type a and b antigens. Lewis double negatives (5-7%) are not able to be characterized as for their secretor status and so will be typed for A, B, and H antigens to allow for phenotypic characterization of all samples. 46

Breath
To interrogate these interventions for changes in SIBO, a pathophysiological process associated with poor linear growth, we will assess hydrogen and methane quantity of breath samples in a subset of 200 participants. Breath samples collected in glass vials will be extracted and analyzed using a Quintron BreathTracker SC breath chromatograph (Quintron Inc, Milwaukee, WI USA). 33 Data on both hydrogen and methane production will be collected.

Data management and confidentiality
Data are collected onto paper forms, which are then quality checked before being double-entered by two different study team data clerks independently at the research site into a secure web-based system (Multi-Schema Information Capture, MuSIC) managed by UVa Department of Public Health Sciences. Any discrepancies in this double-data entry are reconciled by assessing the original paper form. The primary outcome for the overall study is LAZ at 18 months. For each of the intervention domains we will compare this outcome between those randomized to receive the intervention vs. those who did not. This will be performed using a modified intent-to-treat analysis, assessing participants who complete the study and have a measurement of LAZ at 18 months.

Secondary endpoints
Each of the intervention domains has its own set of secondary outcomes as well (see Table 1), and these will be compared between those who receive the intervention and those who do not. This will initially be performed using the same modified intent-to-treat analysis assessing those with the individual outcome of interest. Secondary analysis for all endpoints (including LAZ at 18 months) will also be performed using per-protocol treatment, defined as participants meeting the following criteria: • Breastfed through age 6 months • Received all doses of azithromycin • Received initial dose of nitazoxanide • Received at least 50% of nicotinamide doses as measured by pill-counting.
• Have the outcome of interest measured.

Data and safety monitoring
A data and safety monitoring board has been formed to review data regarding safety of these interventions. This board contains four experts and physicians from Tanzania, the US and the United Kingdom with specialties in gastroenterology, global health, pediatrics, and statistics to assist with reviewing data related to adverse events, health outcomes and surveillance cultures evaluating for antibiotic resistance. Adverse event data are reported and any need to be un-blinded as to treatment allocation or to discontinue any of the treatment arms.

Statistical power
The primary outcome is difference in LAZ at 18 months. The sample size was calculated to provide sufficient power for testing the main effects for the two interventions in a 2 x 2 factorial design. If each main effect were tested at the 5% levels, as is customary in the analysis of factorial studies, 47 48 with 270 participants per group, there is 80% power for a difference in LAZ of 0.176. This is within the range of improvement of an increase of 0.16 seen in a prior study after 12 months of micronutrient supplementation, 49 though in Haydom the growth and deficiencies are more severe than most areas, thus the potential for observing a greater effect size. The interaction between main effects will then be estimated and tested as a secondary analysis. If we were to take a more conservative approach and use a significance level of 2.5% for each of the main effects, with 270 subjects per group, we would still have 80% power to detect an average LAZ increase of 0.193. Finally, if we were to adjust the significance to include the test for the interaction, the F-test for the main effect in a 2-way ANOVA has 80% power with a two-sided significance level of 1.67%, when the main effect of the intervention is to increase the mean LAZ by 0.203. All of these are reasonable changes in LAZ to observe for in the Haydom population, given the considerable demonstrated mean deficit in growth. Adjusting for 10% dropout, the total sample size required is 1188. Within our recruitment radius >2507 births occur yearly. We will recruit during a 12-month window. In our extensive research experience at the Haydom MAL-ED site, we observed 0% refusal to participate and a high 81% retention rate by 4 years of follow-up during MAL-ED (and, thus, a study of shorter duration may have improved follow-up). Based on this, we anticipate a high rate of recruitment (>90%) and 18-month retention (>90%). Therefore, accounting for drop-out, we anticipate achieving 1188  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   21 mother/child dyads divided between intervention and control, leaving us adequate power to determine differences in LAZ. In addition to standard 2-way ANOVA to assess main effects and the interaction, linear and logistic regression will be used to evaluate the effect of the individual intervention domains on outcomes after taking into account individual maternal and child characteristics such as socioeconomic status, maternal height and enrollment weight as covariates.

Study timeline
Enrollment began 5 September 2017 and is planned to continue through 5 September 2018. Children will be followed until age 18 months, resulting in a planned completion in February 2020. Blinding will be maintained until all final study visits are completed. The primary analysis will be performed once data collection is complete for the variables involved.
To reduce intra-assay variability, the majority of laboratory testing will be batched and performed once complete samples are present.

Potential challenges and limitations
Potential challenges include adequate recruitment of target participant numbers within the one-year study window. Should this be true, we will consider extending the recruitment window. Maintaining adherence of study medication, particularly the daily doses of nicotinamide to breast-feeding mothers during months 0-6, and infants during months 6-18, may be a challenge in a setting in which few individuals are accustomed to taking daily medication or nutritional supplements. To protect against adherence challenges, we performed a pilot study beforehand, identifying potential barriers to adherence and resulting in steps to remind mothers frequently during the first month of the study, with follow-up home visits by the field team and by local community healthcare workers.

Dissemination
This trial has been registered at ClinicalTrials.gov (#NCT03268902). Scientific reports resulting from this study will be made publicly available. Following completion of data accrual, data will be made publicly available in an online database. Data will be presented at major conferences such as those of the American Society of Tropical Medicine and Hygiene, the Endocrine Society, and the African Academy of Sciences. Scientific articles will be published with open-access availability at major scientific journals.

Funding statement
This work was supported by the Bill and Melinda Gates Foundation, OPP1141342.

Role of the sponsor and funder
The funder contributed input into the design of the study. Neither the sponsor nor the funder had input into the preparation of this report for publication. The sponsor has ultimate authority into these activities.  Figure 1: The Tryptophan-kynurenine-niacin pathway in the intestinal mucosa. Dietary tryptophan is metabolized to kynurenine by indoleamine 2,3-dioxygenase (IDO)-whose activity is increased in settings of inflammation. Adverse effects of elevated IDO activity relate to modulation of the immune response, decreased activity of mTOR in producing antimicrobial peptides, and decreased tryptophan for protein synthesis. Intervention with nicotinamide ameliorates some of these effects in preclinical models; in the current study, we hypothesize that intervention with nicotinamide will increase linear growth via normalization of the tryptophan-kynurenine-niacin pathway and restored enterocyte activity. Both nicotinic acid and nicotinamide are considered dietary forms of niacin. Abbreviations: IDO = indoleamine 2,3dioxygenase; NAD = nicotinamide adenine dinucleotide.   Figure 1: The Tryptophan-kynurenine-niacin pathway in the intestinal mucosa. Dietary tryptophan is metabolized to kynurenine by indoleamine 2,3-dioxygenase (IDO)-whose activity is increased in settings of inflammation. Adverse effects of elevated IDO activity relate to modulation of the immune response, decreased activity of mTOR in producing antimicrobial peptides, and decreased tryptophan for protein synthesis. Intervention with nicotinamide ameliorates some of these effects in preclinical models; in the current study, we hypothesize that intervention with nicotinamide will increase linear growth via normalization of the tryptophan-kynurenine-niacin pathway and restored enterocyte activity. Both nicotinic acid and nicotinamide are considered dietary forms of niacin. Abbreviations: IDO = indoleamine 2,3dioxygenase; NAD = nicotinamide adenine dinucleotide.
338x190mm (300 x 300 DPI)  Table 1__ Participant timeline 13 Time schedule of enrolment, interventions (including any run-ins and washouts), assessments, and visits for participants. A schematic diagram is highly recommended (see Figure) _____9-10______

Methods: Assignment of interventions (for controlled trials)
Allocation: Sequence generation 16a Method of generating the allocation sequence (eg, computer-generated random numbers), and list of any factors for stratification. To reduce predictability of a random sequence, details of any planned restriction (eg, blocking) should be provided in a separate document that is unavailable to those who enrol participants or assign interventions ______9_______ Allocation concealment mechanism 16b Mechanism of implementing the allocation sequence (eg, central telephone; sequentially numbered, opaque, sealed envelopes), describing any steps to conceal the sequence until interventions are assigned ______9_______ Implementation 16c Who will generate the allocation sequence, who will enrol participants, and who will assign participants to interventions ______9_______ Blinding (masking) 17a Who will be blinded after assignment to interventions (eg, trial participants, care providers, outcome assessors, data analysts), and how If blinded, circumstances under which unblinding is permissible, and procedure for revealing a participant's allocated intervention during the trial ______19-20____

Data collection methods 18a
Plans for assessment and collection of outcome, baseline, and other trial data, including any related processes to promote data quality (eg, duplicate measurements, training of assessors) and a description of study instruments (eg, questionnaires, laboratory tests) along with their reliability and validity, if known. Reference to where data collection forms can be found, if not in the protocol Plans to promote participant retention and complete follow-up, including list of any outcome data to be collected for participants who discontinue or deviate from intervention protocols _____9-10______ Methods for any additional analyses (eg, subgroup and adjusted analyses) _____19_______ 20c Definition of analysis population relating to protocol non-adherence (eg, as randomised analysis), and any statistical methods to handle missing data (eg, multiple imputation) _____19_______

Methods: Monitoring
Data monitoring 21a Composition of data monitoring committee (DMC); summary of its role and reporting structure; statement of whether it is independent from the sponsor and competing interests; and reference to where further details about its charter can be found, if not in the protocol. Alternatively, an explanation of why a DMC is not needed Description of any interim analyses and stopping guidelines, including who will have access to these interim results and make the final decision to terminate the trial _____19-20_____

Harms 22
Plans for collecting, assessing, reporting, and managing solicited and spontaneously reported adverse events and other unintended effects of trial interventions or trial conduct _____19-20_____ Auditing 23 Frequency and procedures for auditing trial conduct, if any, and whether the process will be independent from investigators and the sponsor ______18_______  nicotinamide is administered as a daily oral dose, which for breast-feeding children aged 0-6 months is given to the mother and for children aged 6-18 months is given to the child directly.

Ethics and dissemination
Azithromycin is given to the child as a single oral dose at months 6, 9, 12, and 15; nitazoxanide is given as a 3-day course at months 12 and 15. Mother/child pairs are followed via monthly inhome visits. The primary outcome is the child's length-for-age Z-score at 18 months. Secondary outcomes for the child include additional anthropometry measures; stool pathogen burden and bacterial microbiome; systemic and enteric inflammation; blood metabolomics, growth factors, inflammation and nutrition; hydrogen breath assessment to estimate smallintestinal bacterial overgrowth; and assessment of cognitive development. Secondary outcomes for the mother include breastmilk content of nicotinamide, other vitamins and amino acids; blood measures of tryptophan-kynurenine-niacin pathway; and stool pathogens.
Ethics and dissemination: This trial has been approved by the Tanzanian National Institute for Medical Research, the Tanzanian FDA and the University of Virginia IRB. Findings will be presented at national and international conferences and published in peer-review journals.

STRENGTHS AND LIMITATIONS OF THIS STUDY
• This study uses a factorial design to determine independent contributions of two different intervention approaches onearly childhood growth and development.
• The study is set in a rural site in sub-Saharan Africa with a high degree of both enteric pathogen carriage and stunting.
• This study assesses intervention with nicotinamide, targeting the tryptophan-kynurenineniacin pathway, which may have important roles in intestinal bacterial content and immune and metabolic response.
• In addition to the primary outcome of length-for-age Z-score, this study measures extensive secondary outcomes related to infection, nutrition, metabolism and cognitive development.
• Due to sample size constraints, one limitation of the study is that two different antimicrobial interventions (azithromycin and nitazoxanide) are randomized together, limiting the ability to determine any individual contribution of either of these to study outcomes. worldwide, including the region around Haydom, Tanzania. 3 4 Haydom is in a rural, semi-arid area with a high degree of poverty. Children in Haydom followed in MAL-ED had a prevalence of stunting of 70% at 18 months of age. 5 Data from MAL-ED also showed that linear growth deficits correlated with lower scores for cognitive development, 6 potentially through shared barriers to optimal growth and brain development. 1 7 Antimicrobials to improve childhood growth In MAL-ED, children in the Haydom site had a high burden of infection with intestinal pathogens starting early in life. High carriage of enteropathogens, even in the absence of diarrhea, was associated with a 1.36 increased odds of having lower length for age. 3 As for the specific enteropathogens, among the most prevalent pathogens at age 6 months were enteroaggregative E.coli (48%) and Campylobacter (27%), 8 each of which were associated with a 0.85 cm and 0.83 cm, respectively, shortfall in height attainment. 9 Both of these infections are treatable with azithromycin. Other highly prevalent pathogens by age 6 months in the MAL-ED study included Giardia (22%) and Cryptosporidium (6%), and early Giardia infection was also associated with growth shortfalls. 10 Both of these protozoal infections are treatable with nitazoxanide. These observations raise questions as to whether such antibiotics may be useful in ameliorating growth deficits.
A meta-analysis of 10 randomized controlled trials showed that antibiotic use increased height by 0.04 cm/month. 11 Another study demonstrated that randomized administration of  13 Among children with cryptosporidial infection in Zambia, treatment with nitazoxanide resulted in reduced mortality, while linear growth was not assessed. 14 Therefore, the role of these antimicrobials for linear growth remains unclear.

Nicotinamide to improve childhood growth
Another potential means of improving growth is by targeting the tryptophan-kynurenineniacin pathway ( Figure 1). In this pathway, dietary tryptophan is absorbed and then metabolized by indoleamine 2,3-dioxygenase (IDO) to kynurenine and later to nicotinic acid, a key precursor to NAD+. The importance of this pathway has been supported by recent work identifying mutations in the NAD+ pathway as a cause of birth defects in humans; in animal models with these mutations, birth defects were prevented via supplementation with niacin. 15  Basic science models have demonstrated that tryptophan deficiency is associated with reduced growth velocity. 17 Abnormalities in the tryptophan-kynurenine-niacin pathway are also associated with intestinal inflammation, including during response to intestinal pathogens. 18 19 In animal models, an inability to transport tryptophan into enterocytes results in diarrhea, while provision of nicotinamide (a form of niacin) partly restores function of mTOR in enterocytes, increasing secretion of antimicrobial peptides and reducing diarrhea, 18 19 suggesting effects on diarrheal  20 21 and during infections; 22 this appears to be a maladaptive in most cases, as kynurenine has potential immunosuppressive effects. 23 24 In MAL-ED, children at the Haydom site with higher ratios of kynurenine:tryptophan were noted to have poorer linear growth. 24 Given that 1) children in the Haydom area have a maize-based diet low in tryptophan, a key precursor to niacin/nicotinamide and NAD+, 2) preclinical studies demonstrate improvements in antimicrobial peptides and reduced diarrhea with nicotinamide supplementation, 3) an increase in nicotinamide as a precursor of NAD+ should cause reduced need for the tryptophankynurenine-niacin pathway, reducing activity of IDO and thereby reducing kynurenine, and 4) higher kynurenine:tryptophan is associated with poor growth, we reasoned that intervention with nicotinamide among children in this region would improve linear growth. However, because growth is a long-term process and nicotinamide is a water-soluble vitamin without a mechanism for on-going storage, we reasoned that an intervention with nicotinamide would need to be given over a long period of time.

Study hypotheses
We hypothesized that among children with a high degree of stunting and enteric pathogen burden in this region, intervention with 1) antimicrobials and/or 2) nicotinamide would increase linear growth. We further set out to interrogate the potential mechanisms behind these effects. In the case of antimicrobial intervention, we hypothesize that intervention with azithromycin and nitazoxanide (relative to placebo) would result in a decrease in enteropathogen burden that would be associated with an increase in growth. In the case of nicotinamide (relative to placebo), we hypothesize that intervention will result in a reduction in the kynurenine:tryptophan ratio, an increase in availability of nicotinamide for NAD+ synthesis and associated increases in growth. are to determine if interventions with 1) antimicrobials and 2) nicotinamide are associated with increased length-for-age z-score (LAZ, relative to placebo-treated children) by age 18 months.
Specifically, the antimicrobial intervention consists of azithromycin administered as a single dose at age 6, 9, 12, and 15 months and nitazoxanide as a 3-day course at age 12 and 15 months (with both antimicrobials randomized together). The nicotinamide intervention consists of nicotinamide 250 mg daily administered daily by mouth to the child's mother during breast feeding 0-6 months (and thus passed on to the child via breast milk 25

Eligibility
Participants are eligible if the mother is >18 years and the child is age <14 days and was born in the Haydom catchment area (defined for study purposes as 25 km radius of Haydom Lutheran Hospital).
Exclusion criteria are as follows: maternal inability to adhere to the protocol, multiple gestation, significant birth defect or neonatal illness, weight <1,500 g at enrollment (age < 14 days), lack of intent to breastfeed infant and plan to move from area within the next 18 months.

Recruitment
Recruitment occurs via survey by field teams of surrounding villages and via input from local community healthcare workers. Pregnant women are identified and informed regarding the study. Pregnant women in the third trimester may provide informed consent for the study but are not enrolled until the initial assessment visit of the infant to determine eligibility (which includes ascertainment of infant weight). If informed consent is not obtained during pregnancy, it is obtained at enrollment. The general timeline of study procedures is shown in Table 1. which should be 1.5 or above. Informed consent is provided at this time, if it was not obtained during pregnancy. Eligibility (besides current weight) is confirmed and the mother/child pair is randomized in the study.

Randomization
Randomization occurs at the time of enrollment. Block randomization was used in permuted blocks of eight for the four intervention combinations (antimicrobial + nicotinamide, antimicrobial alone, nicotinamide alone, all placebo). Study teams are assigned permuted blocks before initial study visits to sequentially allocate randomization status to participants based on study identification number assigned prior to the initial study visit.

Intervention
As a factorial study design, the two intervention domains (antimicrobial and nicotinamide) are randomized separately. The antimicrobial intervention consists of two medications that are randomized together: azithromycin and nitazoxanide (vs. placebos of both Haydom-specific data from MAL-ED demonstrated a high prevalence (96%) of breast feeding until children were 6 months old. 5 Because the prevalence of exclusive breastfeeding was low in the area around Haydom during MAL-ED, the current study includes a breast-feeding encouragement and support initiative, utilizing nurses who have received specialized training in breast feeding; these nurse work with the local community healthcare workers in home visits to participating families in the first 6 months of the study. Their message is that exclusive breast feeding is better for mother and baby (highlighting a variety of reasons) and that if the family chooses not to breast feed, that providing breast milk before any complimentary foods will allow the child to receive a greater amount of the needed nutrition-including a greater amount of the nicotinamide ingested by the mother. The lack of intent to breast feed is an exclusion criterion at the time of enrollment; however, if the mother dies or discontinues breast feeding, the child subsequently would not receive nicotinamide/placebo until age 6 months.
From months 6-18 the child receives sachets of nicotinamide 100 mg by mouth daily.
Mothers are instructed to mix the nicotinamide powder with a small volume of age-appropriate food (e.g. infant cereal) to give to the child. For both the maternal nicotinamide pills and the children's nicotinamide sachets, doses are given as a 60-day supply every two months. When a new supply is provided, the family is asked for the remainder of doses not consumed the prior 60 days. These unused doses are counted and recorded by pharmacy staff to determine adherence.

Questionnaires
Study staff use standardized questionnaires designed for the study, in many cases adapted from questionnaires used in the MAL-ED study. These include questions asked at each monthly visit regarding breastfeeding, use of complementary foods, childhood illness and treatment (including whether an antibiotic was used and if so which one), types of food sources given, and food insecurity. Additional questionnaires are asked at other specific time periods, including aspects of the child's home environment (completed at the first month's visit) and the vaccines that the child has received (completed at the 12-month visit).

Anthropometry measures
All field team members are trained in the accurate measurement of anthropometry indices. The child's weight is measured at the initial visit and at each monthly visit through age 18 months. The child's length and head circumference are measured at the initial visit and every three visits thereafter through 18 month. The mother's weight and height are measured at month 10-11. Quarterly, weights are measured using a digital scale that undergoes quality assessments weekly. Infants are weighed on an infant tray with the infant nude or wearing a dry diaper or standing, when able. In between these quarterly weight measures, weights are measured monthly using hanging spring scales that are zeroed before each use and assessed for accuracy weekly. Lengths are measured quarterly by two field team members with the child lying flat on a measuring board with a fixed, perpendicular board against which the head is

Developmental assessments
To assess maternal-child interaction, an important predictor of early childhood development and health, 26 27 The Observation of Maternal Child Interaction (OMCI) 28 assessment is collected at 12 months of age. This observation includes a 5 minute observation of mother and child interacting over a picture book, with ratings given for positive affect, verbal statements, sensitivity, scaffolding, language stimulation, focus and mutual enjoyment, among others.
At 18 months, developmental assessment is performed by trained assessors using two standardized scores, the Malawi Developmental Assessment Tool (MDAT) 29  and caregiver report of skills go into the score.
The Bayley cognitive assessment involves a series of tasks presented to the infant, which assess for play, reasoning, and mental representation. This is a commonly used assessment for early childhood and has been used in many research sites around the world. 6 31 Specimen collection and place inside a transport box with cold packs. The stool collection will be timed prior to a planned study visit, such that the field-worker can collect the sample and the laboratory process the specimen shortly thereafter. Stool samples are then stored at -80 degrees. A subset of specimens, prior to freezing will be cultured for E.coli and screened for antibiotic resistance

Rationale
During the first 6 months of participation, the only means of the child receiving intervention is through maternal delivery of nicotinamide in breast milk. However, nicotinamide in vivo is both derived from tryptophan and converted to multiple other vitamers. 34 As such, we will assess content of nicotinamide, vitamers and tryptophan in breast milk at two time points during the intervention.
Maternal breast milk samples collected at months 1 and 5 will be tested at the United States Department of Agriculture laboratory at the University of California -Davis. Planned LC-MS analyses includes nicotinamide, thiamin, riboflavin, flavin adenine dinucleotide, pyridoxal, pyridoxine, biotin and pantothenic acid concentrations, and on a random subset (n=600 samples) niacin vitamer, and free and total tryptophan concentrations.

Rationale
Blood testing will include assessment of multiple measures both to verify safety of the interventions and interrogate the effects of the intervention on physiological processes related to linear growth. Safety assessment includes testing of general tests of health with basic metabolic panel, liver function and enzyme testing, and complete blood count. This will assess for differences between intervention groups in measures of kidney, liver and hematopoetic health. Additional tests to interrogate the physiologic effects of the intervention include  Collagen X: We will assess for intervention-associated changes in collagen X (a biomarker of rapid skeletal growth) using an ELISA developed at Shriners Hospital for Children in Portland, OR, and Oregon Health and Science University. 36 Metabolomics: The tryptophan-kynurenine-niacin pathway will be assessed at 12 and 18 months via metabolomics analyses of serum and plasma using 1 H nuclear magnetic resonance

Stool
Stool testing will include assessment of the following: Microbiome: Stool bacterial content will be assessed at in a subset of 200 participants at the University of Maryland. Total stool DNA will be isolated by bead beating method. A library will be created each specimen. The libraries will be multiplexed and sequenced by Illumina HiSeq.
The resulting reads will be cleaned, binned by specimen ID, separated in operational taxonomic units (OTUs) by DNAclust 39 and compared using metagenomeseq. Metagenomeseq normalizes the data and introduces a correction factor to account for under-sampling of OTUs.
The normalized data will be compared using odds ratios and multiple t-tests, with p-values adjusted for multiple testing.
Taqman Array Cards (TAC): Pathogen burden will be assessed using TAC cards ( Figure 2) as performed previously. 40  Antibiotic resistance: We will perform stool bacterial culture on a subset of participants to monitor for development of antimicrobial resistance, which is a valid concern of any mass antimicrobial administration. Briefly, stool will be cultured on MacConkey agar and 5 lactose fermenting E. coli colonies pooled and tested for AMR using disc diffusion to screen for ESBL, AmpC, carbapenemase and susceptibility to antibiotics commonly in use locally.
Azrythromycin susceptibility will be tested by E-Test. 43 Myeloperoxidase: As an assessment of stool inflammation, myeloperoxidase will be assessed at months 6, 12 and 18. Testing will be performed at the Global Health Research Center laboratory in Haydom using a commercially-available assay (Alpco, Salem, NH).

Urine
Urine samples at 6, 12 and 18 months will assessed alongside serum and plasma samples at Imperial College, London for untargeted metabolomics analysis by NMR spectroscopy and UPLC-MS, as described above. 38

Saliva
For a more complete assessment of these interventions on intestinal inflammation in this setting, we will test mothers and children for secretor status of fucosyltransferase 2 (FUT2), a regulator of blood group antigens on the intestinal mucosa, as non-secretor status has been associated with infections 44 and other forms of colitis. 45 DNA for mothers and children will be extracted from salivary samples at UVa and FUT2 genotype will be determined by dot blot assays in a hierarchical method based largely on Lewis Type a and b antigens. Lewis double

Breath
To interrogate these interventions for changes in SIBO, a pathophysiological process associated with poor linear growth, we will assess hydrogen and methane quantity of breath samples in a subset of 200 participants. Breath samples collected in glass vials will be extracted and analyzed using a Quintron BreathTracker SC breath chromatograph (Quintron Inc, Milwaukee, WI USA). 33 Data on both hydrogen and methane production will be collected.

Data management and confidentiality
Data are collected onto paper forms, which are then quality checked before being double-entered by two different study team data clerks independently at the research site into a secure web-based system (Multi-Schema Information Capture, MuSIC) managed by UVa Department of Public Health Sciences. Any discrepancies in this double-data entry are reconciled by assessing the original paper form.

Primary endpoint
The primary outcome for the overall study is LAZ at 18 months. For each of the intervention domains we will compare this outcome as a t-test between those randomized to receive the intervention vs. those who did not. This will be performed using a modified intent-totreat analysis, assessing participants who complete the study and have a measurement of LAZ at 18 months. Each of the intervention domains has its own set of secondary outcomes as well (see Table 2), and these will be compared between those who receive the intervention and those who do not. This will initially be performed using the same modified intent-to-treat analysis assessing those with the individual outcome of interest. This will be performed as t-tests for continuous outcomes and chi square tests for binary outcomes.

Secondary endpoints
Secondary analysis for all endpoints (including LAZ at 18 months) will also be performed using per-protocol treatment, defined as participants meeting the following criteria: • Breastfed through age 6 months • Received all doses of azithromycin • Received initial dose of nitazoxanide • Received at least 50% of nicotinamide doses as measured by pill-and sachet counting.
• Have the outcome of interest measured.

Blinding
The study is being performed in a double-blinded manner. Individual study medications were purchased from the same manufacturer as their corresponding placebo, and placebos were designed to be similar in appearance and taste. Study personnel received the medications and placebos merely labeled as "A" and "B," with the key to this allocation provided to one individual at UVa not otherwise associated with the study. Blinding will be maintained throughout the study unless adverse events occur in significantly disproportionate amounts by intervention allocation status. Allocation status will otherwise only be revealed following completion of collection of data and samples. recommendations regarding ongoing safety of the interventions and any need to be un-blinded due to differences in adverse events between treatment allocation group and can recommend discontinuation of any of the intervention arms.

Statistical power
The primary outcome is difference between intervention groups in LAZ at 18 months.
The sample size was calculated to provide sufficient power for testing the main effects using separate t-tests for each of the two interventions in this 2 x 2 factorial design, using a standard deviation of 1.03 HAZ at 18 months. If each main effect were tested at the 5% levels, as is customary in the analysis of factorial studies, 47 48 with 270 participants per group, there is 80% power for a difference in LAZ of 0.176. This is within the range of improvement of an increase of 0.16 seen in a prior study after 12 months of micronutrient supplementation, 49

though in
Haydom the growth and deficiencies are more severe than most areas, thus the potential for observing a greater effect size. The interaction between main effects of these interventions will then be estimated and tested as a secondary analysis using a 2-way ANOVA. If we were to take a more conservative approach and use a significance level of 2.5% for each of the main effects, with 270 subjects per group, we would still have 80% power to detect an average LAZ increase of 0.193. Finally, if we were to adjust the significance to include the test for the interaction, the F-test for the main effect in a 2-way ANOVA has 80% power with a two-sided significance level of 1.67%, when the main effect of the intervention is to increase the mean LAZ

Statistics
In addition to use of ITT for the central analysis (t-tests) and the standard 2-way ANOVA to assess main effects and the interaction, we will perform several sensitivity analyses to assess for influences from children who dropped out or died during the study. These will include assessments of last-measurement carried forward and assumptions about those who dropped out being taller or shorter than the mean. Additional secondary analyses will include assessing hazard ratios for events such as mortality and hospitalization, including time to event analysis.

Study timeline
Enrollment began 5 September 2017 and is planned to continue through 5 September 2018. Children will be followed until age 18 months, resulting in a planned completion in February 2020. Blinding will be maintained until all final study visits are completed. The primary analysis will be performed once data collection is complete for the variables involved.
To reduce intra-assay variability, the majority of laboratory testing will be batched and performed once complete samples are present.

Potential challenges and limitations
Potential challenges include adequate recruitment of target participant numbers within the one-year study window. Should this be true, we will consider extending the recruitment window. Maintaining adherence of study medication, particularly the daily doses of nicotinamide to breast-feeding mothers during months 0-6, and infants during months 6-18, may be a challenge in a setting in which few individuals are accustomed to taking daily medication or nutritional supplements. To protect against adherence challenges, we performed a pilot study beforehand, identifying potential barriers to adherence and resulting in steps to remind mothers frequently during the first month of the study, with follow-up home visits by the field team and by local community healthcare workers.

Regulatory authorities
ELICIT has received approval from Tanzania  Patients and the public were not directly involved in the planning of this study. Results will be communicated to participants via community meetings with local leaders.

Dissemination
This trial has been registered at ClinicalTrials.gov (#NCT03268902). Scientific reports resulting from this study will be made publicly available. Following completion of data accrual, data will be made publicly available in an online database. Data will be presented at major conferences such as those of the American Society of Tropical Medicine and Hygiene, the Pediatric Academic Society, the Endocrine Society, and the African Academy of Sciences.
Scientific articles will be published with open-access availability at major scientific journals.

Role of the sponsor and funder
The funder contributed input into the design of the study. Neither the sponsor nor the funder had input into the preparation of this report for publication. The sponsor has ultimate authority into these activities.

Competing interests statement
None of the authors or study co-investigators have any competing interests to declare.

Figure 1: The Tryptophan-kynurenine-niacin pathway in the intestinal mucosa.
Dietary tryptophan is metabolized to kynurenine by indoleamine 2,3-dioxygenase (IDO)-whose activity is increased in settings of inflammation. Adverse effects of elevated IDO activity relate to modulation of the immune response, decreased activity of mTOR in producing antimicrobial peptides, and decreased tryptophan for protein synthesis. Intervention with nicotinamide ameliorates some of these effects in preclinical models; in the current study, we hypothesize that intervention with nicotinamide will increase linear growth via normalization of the tryptophan-kynurenine-niacin pathway and restored enterocyte activity. Both nicotinic acid and nicotinamide are considered dietary forms of niacin. Abbreviations: IDO = indoleamine 2,3dioxygenase; NAD = nicotinamide adenine dinucleotide.      Figure 1: The Tryptophan-kynurenine-niacin pathway in the intestinal mucosa. Dietary tryptophan is metabolized to kynurenine by indoleamine 2,3-dioxygenase (IDO)-whose activity is increased in settings of inflammation. Adverse effects of elevated IDO activity relate to modulation of the immune response, decreased activity of mTOR in producing antimicrobial peptides, and decreased tryptophan for protein synthesis. Intervention with nicotinamide ameliorates some of these effects in preclinical models; in the current study, we hypothesize that intervention with nicotinamide will increase linear growth via normalization of the tryptophan-kynurenine-niacin pathway and restored enterocyte activity. Both nicotinic acid and nicotinamide are considered dietary forms of niacin. Abbreviations: IDO = indoleamine 2,3dioxygenase; NAD = nicotinamide adenine dinucleotide.
338x190mm (300 x 300 DPI)    Table 1__ Participant timeline 13 Time schedule of enrolment, interventions (including any run-ins and washouts), assessments, and visits for participants. A schematic diagram is highly recommended (see Figure) _____9-10______ Definition of analysis population relating to protocol non-adherence (eg, as randomised analysis), and any statistical methods to handle missing data (eg, multiple imputation) _____19_______

Methods: Monitoring
Data monitoring 21a Composition of data monitoring committee (DMC); summary of its role and reporting structure; statement of whether it is independent from the sponsor and competing interests; and reference to where further details about its charter can be found, if not in the protocol. Alternatively, an explanation of why a DMC is not needed Description of any interim analyses and stopping guidelines, including who will have access to these interim results and make the final decision to terminate the trial  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  nicotinamide is administered as a daily oral dose, which for breast-feeding children aged 0-6 months is given to the mother and for children aged 6-18 months is given to the child directly.
Azithromycin is given to the child as a single oral dose at months 6, 9, 12, and 15; nitazoxanide is given as a 3-day course at months 12 and 15. Mother/child pairs are followed via monthly inhome visits. The primary outcome is the child's length-for-age Z-score at 18 months. Secondary outcomes for the child include additional anthropometry measures; stool pathogen burden and bacterial microbiome; systemic and enteric inflammation; blood metabolomics, growth factors, inflammation and nutrition; hydrogen breath assessment to estimate smallintestinal bacterial overgrowth; and assessment of cognitive development. Secondary outcomes for the mother include breastmilk content of nicotinamide, other vitamins and amino acids; blood measures of tryptophan-kynurenine-niacin pathway; and stool pathogens.
Ethics and dissemination: This trial has been approved by the Tanzanian National Institute for Medical Research, the Tanzanian FDA and the University of Virginia IRB. Findings will be presented at national and international conferences and published in peer-review journals.

STRENGTHS AND LIMITATIONS OF THIS STUDY
• This study uses a factorial design to determine independent contributions of two different intervention approaches onearly childhood growth and development.
• The study is set in a rural site in sub-Saharan Africa with a high degree of both enteric pathogen carriage and stunting.
• This study assesses intervention with nicotinamide, targeting the tryptophan-kynurenineniacin pathway, which may have important roles in intestinal bacterial content and immune and metabolic response.
• In addition to the primary outcome of length-for-age Z-score, this study measures extensive secondary outcomes related to infection, nutrition, metabolism and cognitive development.
• Due to sample size constraints, one limitation of the study is that two different antimicrobial interventions (azithromycin and nitazoxanide) are randomized together, limiting the ability to determine any individual contribution of either of these to study outcomes.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  worldwide, including the region around Haydom, Tanzania. 3 4 Haydom is in a rural, semi-arid area with a high degree of poverty. Children in Haydom followed in MAL-ED had a prevalence of stunting of 70% at 18 months of age. 5 Data from MAL-ED also showed that linear growth deficits correlated with lower scores for cognitive development, 6 potentially through shared barriers to optimal growth and brain development. 1 7 Antimicrobials to improve childhood growth In MAL-ED, children in the Haydom site had a high burden of infection with intestinal pathogens starting early in life. High carriage of enteropathogens, even in the absence of diarrhea, was associated with a 1.36 increased odds of having lower length for age. 3 As for the specific enteropathogens, among the most prevalent pathogens at age 6 months were enteroaggregative E.coli (48%) and Campylobacter (27%), 8 each of which were associated with a 0.85 cm and 0.83 cm, respectively, shortfall in height attainment. 9 Both of these infections are treatable with azithromycin. Other highly prevalent pathogens by age 6 months in the MAL-ED study included Giardia (22%) and Cryptosporidium (6%), and early Giardia infection was also associated with growth shortfalls. 10 Both of these protozoal infections are treatable with nitazoxanide. These observations raise questions as to whether such antibiotics may be useful in ameliorating growth deficits.
A meta-analysis of 10 randomized controlled trials showed that antibiotic use increased height by 0.04 cm/month. 11 Another study demonstrated that randomized administration of  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   6 azithromycin to asymptomatic young children in Ethiopia was associated with a halving of mortality, 12 suggesting overall effects of suppressing bacterial pathogens in areas with high rates of infectious disease. However results from individual studies of azithromycin mass drugadministration on growth have not shown consistent results. 13 Among children with cryptosporidial infection in Zambia, treatment with nitazoxanide resulted in reduced mortality, while linear growth was not assessed. 14 Therefore, the role of these antimicrobials for linear growth remains unclear.

Nicotinamide to improve childhood growth
Another potential means of improving growth is by targeting the tryptophan-kynurenineniacin pathway (Figure 1). In this pathway, dietary tryptophan is absorbed and then metabolized by indoleamine 2,3-dioxygenase (IDO) to kynurenine and later to nicotinic acid, a key precursor to NAD+. The importance of this pathway has been supported by recent work identifying mutations in the NAD+ pathway as a cause of birth defects in humans; in animal models with these mutations, birth defects were prevented via supplementation with niacin. 15  Basic science models have demonstrated that tryptophan deficiency is associated with reduced growth velocity. 17 Abnormalities in the tryptophan-kynurenine-niacin pathway are also associated with intestinal inflammation, including during response to intestinal pathogens. 18 19 In animal models, an inability to transport tryptophan into enterocytes results in diarrhea, while provision of nicotinamide (a form of niacin) partly restores function of mTOR in enterocytes, increasing secretion of antimicrobial peptides and reducing diarrhea, 18 19 suggesting effects on diarrheal  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   7 pathogens. IDO activity is increased during inflammation, including in inflammatory bowel disease 20 21 and during infections; 22 this appears to be a maladaptive in most cases, as kynurenine has potential immunosuppressive effects. 23 24 In MAL-ED, children at the Haydom site with higher ratios of kynurenine:tryptophan were noted to have poorer linear growth. 24 Given that 1) children in the Haydom area have a maize-based diet low in tryptophan, a key precursor to niacin/nicotinamide and NAD+, 2) preclinical studies demonstrate improvements in antimicrobial peptides and reduced diarrhea with nicotinamide supplementation, 3) an increase in nicotinamide as a precursor of NAD+ should cause reduced need for the tryptophankynurenine-niacin pathway, reducing activity of IDO and thereby reducing kynurenine, and 4) higher kynurenine:tryptophan is associated with poor growth, we reasoned that intervention with nicotinamide among children in this region would improve linear growth. However, because growth is a long-term process and nicotinamide is a water-soluble vitamin without a mechanism for on-going storage, we reasoned that an intervention with nicotinamide would need to be given over a long period of time.

Eligibility
Participants are eligible if the mother is >18 years and the child is age <14 days and was born in the Haydom catchment area (defined for study purposes as 25 km radius of Haydom Lutheran Hospital).
Exclusion criteria are as follows: maternal inability to adhere to the protocol, multiple gestation, significant birth defect or neonatal illness, weight <1,500 g at enrollment (age < 14 days), lack of intent to breastfeed infant and plan to move from area within the next 18 months.

Recruitment
Recruitment occurs via survey by field teams of surrounding villages and via input from local community healthcare workers. Pregnant women are identified and informed regarding the study. Pregnant women in the third trimester may provide informed consent for the study but are not enrolled until the initial assessment visit of the infant to determine eligibility (which includes ascertainment of infant weight). If informed consent is not obtained during pregnancy, it is obtained at enrollment. The general timeline of study procedures is shown in Table 1.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o n l y

Enrollment
Infant/mother pairs are enrolled during the initial study visit to their homes between birth and day of life 14, after eligibility criteria are confirmed, including measurement of infant weight which should be 1.5 or above. Informed consent is provided at this time, if it was not obtained during pregnancy. Eligibility (besides current weight) is confirmed and the mother/child pair is randomized in the study.

Randomization
Randomization occurs at the time of enrollment. Block randomization was used in permuted blocks of eight for the four intervention combinations (antimicrobial + nicotinamide, antimicrobial alone, nicotinamide alone, all placebo). Study teams are assigned permuted blocks before initial study visits to sequentially allocate randomization status to participants based on study identification number assigned prior to the initial study visit.

Intervention
As a factorial study design, the two intervention domains (antimicrobial and nicotinamide) are randomized separately. The antimicrobial intervention consists of two medications that are randomized together: azithromycin and nitazoxanide (vs. placebos of both Haydom-specific data from MAL-ED demonstrated a high prevalence (96%) of breast feeding until children were 6 months old. 5 Because the prevalence of exclusive breastfeeding was low in the area around Haydom during MAL-ED, the current study includes a breast-feeding encouragement and support initiative, utilizing nurses who have received specialized training in breast feeding; these nurse work with the local community healthcare workers in home visits to participating families in the first 6 months of the study. Their message is that exclusive breast feeding is better for mother and baby (highlighting a variety of reasons) and that if the family chooses not to breast feed, that providing breast milk before any complimentary foods will allow the child to receive a greater amount of the needed nutrition-including a greater amount of the nicotinamide ingested by the mother. The lack of intent to breast feed is an exclusion criterion at the time of enrollment; however, if the mother dies or discontinues breast feeding, the child subsequently would not receive nicotinamide/placebo until age 6 months.
From months 6-18 the child receives sachets of nicotinamide 100 mg by mouth daily.
Mothers are instructed to mix the nicotinamide powder with a small volume of age-appropriate food (e.g. infant cereal) to give to the child. For both the maternal nicotinamide pills and the children's nicotinamide sachets, doses are given as a 60-day supply every two months. When a new supply is provided, the family is asked for the remainder of doses not consumed the prior 60 days. These unused doses are counted and recorded by pharmacy staff to determine adherence.

Questionnaires
Study staff use standardized questionnaires designed for the study, in many cases adapted from questionnaires used in the MAL-ED study. These include questions asked at each monthly visit regarding breastfeeding, use of complementary foods, childhood illness and treatment (including whether an antibiotic was used and if so which one), types of food sources given, and food insecurity. Additional questionnaires are asked at other specific time periods, including aspects of the child's home environment (completed at the first month's visit) and the vaccines that the child has received (completed at the 12-month visit).

Anthropometry measures
All field team members are trained in the accurate measurement of anthropometry indices. The child's weight is measured at the initial visit and at each monthly visit through age 18 months. All anthropometry measures for a given month (including the final visit) are able to be performed up to 14 days before or after the target date (which is based on the number of months after the child's birthday. The child's length and head circumference are measured at the initial visit and every three visits thereafter through 18 month. The mother's weight and height are measured at month 10-11. Quarterly, weights are measured using a digital scale that undergoes quality assessments weekly. Infants are weighed on an infant tray with the infant nude or wearing a dry diaper or standing, when able. In between these quarterly weight measures, weights are measured monthly using hanging spring scales that are zeroed before if these are not within 2 mm, a third measurement is obtained and the average of the two closest measurements is used. Head circumference is measured quarterly by placing a nondistensible tape around the child's head just above the eyebrows, above the ears, and around the biggest part of the back of the head. Mid-upper arm circumference is measured quarterly by using the same measuring tape at the mid-point of the upper arm segment. Maternal weight is measured using a standing scale. Maternal height is measured using a portable stadiometer.
For quality assurance, 5% of infant weights, lengths, and head circumferences are measured by a supervisor skilled in measurement to compare to the field team's measurement.

Developmental assessments
To assess maternal-child interaction, an important predictor of early childhood development and health, 26 27 The Observation of Maternal Child Interaction (OMCI) 28 assessment is collected at 12 months of age. This observation includes a 5 minute observation of mother and child interacting over a picture book, with ratings given for positive affect, verbal statements, sensitivity, scaffolding, language stimulation, focus and mutual enjoyment, among others.
At 18 months, developmental assessment is performed by trained assessors using two standardized scores, the Malawi Developmental Assessment Tool (MDAT) 29

and the Bayley
Scales of Infant and Toddler Development, cognitive assessment (Bayley). 30 The MDAT is a validated score that tests multiple childhood development domains, including gross motor, fine motor, language and social development that has been validated in this age range in Malawi and intended for use in rural sub-Saharan settings. Briefly, this assessment was designed to be and caregiver report of skills go into the score.
The Bayley cognitive assessment involves a series of tasks presented to the infant, which assess for play, reasoning, and mental representation. This is a commonly used assessment for early childhood and has been used in many research sites around the world. 6 31 Specimen collection  specimens, prior to freezing will be cultured for E.coli and screened for antibiotic resistance (AMR). This is performed as surveillance for safety and will be performed on a randomlyselected 25 participants per intervention allocation group at 12 and 15 months.
Urine (months 6, 12, and 18): Urine is collected by placing a bag over the child's genitalia at the beginning of a study visit. The child is given adequate hydration, via breast feeding if possible.
Collected urine is then transferred in a sterile container that is stored on ice prior to transport to the research site and then stored at -80 degrees.

Rationale
During the first 6 months of participation, the only means of the child receiving intervention is through maternal delivery of nicotinamide in breast milk. However, nicotinamide in vivo is both derived from tryptophan and converted to multiple other vitamers. 34 As such, we will assess content of nicotinamide, vitamers and tryptophan in breast milk at two time points during the intervention.
Maternal breast milk samples collected at months 1 and 5 will be tested at the United States Department of Agriculture laboratory at the University of California -Davis. Planned LC-MS analyses includes nicotinamide, thiamin, riboflavin, flavin adenine dinucleotide, pyridoxal, pyridoxine, biotin and pantothenic acid concentrations, and on a random subset (n=600 samples) niacin vitamer, and free and total tryptophan concentrations.

Rationale
Blood testing will include assessment of multiple measures both to verify safety of the interventions and interrogate the effects of the intervention on physiological processes related to linear growth. Safety assessment includes testing of general tests of health with basic metabolic panel, liver function and enzyme testing, and complete blood count. This will assess for differences between intervention groups in measures of kidney, liver and hematopoetic health. Additional tests to interrogate the physiologic effects of the intervention include  Insulin-like growth factor-1 (IGF-1) and high-sensitivity C-reactive protein (hsCRP): To interrogate these interventions for changes in growth factor and low-grade systemic inflammation, respectively, IGF-1 and hsCRP will be tested at 12 and 18 months at the UVa Center for Research in Reproduction Ligand Core Laboratory using an Immulite 2000.
Collagen X: We will assess for intervention-associated changes in collagen X (a biomarker of rapid skeletal growth) using an ELISA developed at Shriners Hospital for Children in Portland, OR, and Oregon Health and Science University. 36 Metabolomics: The tryptophan-kynurenine-niacin pathway will be assessed at 12 and 18 months via metabolomics analyses of serum and plasma using 1 H nuclear magnetic resonance

Stool
Stool testing will include assessment of the following: Microbiome: Stool bacterial content will be assessed at in a subset of 200 participants at the University of Maryland. Total stool DNA will be isolated by bead beating method. A library will be created each specimen. The libraries will be multiplexed and sequenced by Illumina HiSeq.
The resulting reads will be cleaned, binned by specimen ID, separated in operational taxonomic units (OTUs) by DNAclust 39 and compared using metagenomeseq. Metagenomeseq normalizes the data and introduces a correction factor to account for under-sampling of OTUs.
The normalized data will be compared using odds ratios and multiple t-tests, with p-values adjusted for multiple testing.
Azrythromycin susceptibility will be tested by E-Test. 43 Myeloperoxidase: As an assessment of stool inflammation, myeloperoxidase will be assessed at months 6, 12 and 18. Testing will be performed at the Global Health Research Center laboratory in Haydom using a commercially-available assay (Alpco, Salem, NH).

Urine
Urine samples at 6, 12 and 18 months will assessed alongside serum and plasma samples at Imperial College, London for untargeted metabolomics analysis by NMR spectroscopy and UPLC-MS, as described above. 38

Breath
To interrogate these interventions for changes in SIBO, a pathophysiological process associated with poor linear growth, we will assess hydrogen and methane quantity of breath samples in a subset of 200 participants. Breath samples collected in glass vials will be extracted and analyzed using a Quintron BreathTracker SC breath chromatograph (Quintron Inc, Milwaukee, WI USA). 33 Data on both hydrogen and methane production will be collected.

Data management and confidentiality
Data are collected onto paper forms, which are then quality checked before being double-entered by two different study team data clerks independently at the research site into a secure web-based system (Multi-Schema Information Capture, MuSIC) managed by UVa Department of Public Health Sciences. Any discrepancies in this double-data entry are reconciled by assessing the original paper form.

Primary endpoint
The primary outcome for the overall study is LAZ at 18 months. For each of the intervention domains we will compare this outcome as a t-test between those randomized to receive the intervention vs. those who did not. This will be performed using a modified intent-totreat analysis, assessing participants who complete the study and have a measurement of LAZ at 18 months.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   21 Each of the intervention domains has its own set of secondary outcomes as well (see Table 2), and these will be compared between those who receive the intervention and those who do not. This will initially be performed using the same modified intent-to-treat analysis assessing those with the individual outcome of interest. This will be performed as t-tests for continuous outcomes and chi square tests for binary outcomes.

Secondary endpoints
Secondary analysis for all endpoints (including LAZ at 18 months) will also be performed using per-protocol treatment, defined as participants meeting the following criteria: • Breastfed through age 6 months • Received all doses of azithromycin • Received initial dose of nitazoxanide • Received at least 50% of nicotinamide doses as measured by pill-and sachet counting.
• Have the outcome of interest measured.

Blinding
The study is being performed in a double-blinded manner. Individual study medications were purchased from the same manufacturer as their corresponding placebo, and placebos were designed to be similar in appearance and taste. Study personnel received the medications and placebos merely labeled as "A" and "B," with the key to this allocation provided to one individual at UVa not otherwise associated with the study. Blinding will be maintained throughout the study unless adverse events occur in significantly disproportionate amounts by intervention allocation status. Allocation status will otherwise only be revealed following completion of collection of data and samples.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  recommendations regarding ongoing safety of the interventions and any need to be un-blinded due to differences in adverse events between treatment allocation group and can recommend discontinuation of any of the intervention arms.

Statistical power
The primary outcome is difference between intervention groups in LAZ at 18 months.
The sample size was calculated to provide sufficient power for testing the main effects using separate t-tests for each of the two interventions in this 2 x 2 factorial design, using a standard deviation of 1.03 HAZ at 18 months. If each main effect were tested at the 5% levels, as is customary in the analysis of factorial studies, 47 48 with 270 participants per group, there is 80% power for a difference in LAZ of 0.176. This is within the range of improvement of an increase of 0.16 seen in a prior study after 12 months of micronutrient supplementation, 49

though in
Haydom the growth and deficiencies are more severe than most areas, thus the potential for observing a greater effect size. The interaction between main effects of these interventions will then be estimated and tested as a secondary analysis using a 2-way ANOVA. If we were to take a more conservative approach and use a significance level of 2.5% for each of the main effects, with 270 subjects per group, we would still have 80% power to detect an average LAZ increase of 0.193. Finally, if we were to adjust the significance to include the test for the interaction, the F-test for the main effect in a 2-way ANOVA has 80% power with a two-sided significance level of 1.67%, when the main effect of the intervention is to increase the mean LAZ  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59

Statistics
In addition to use of ITT for the central analysis (t-tests) and the standard 2-way ANOVA to assess main effects and the interaction, we will perform several sensitivity analyses to assess for influences from children who dropped out or died during the study. These will include use of linear mixed-effects models based on repeated measures of anthropometric measures (adjusted for baseline LAZ), assessments of last-measurement carried forward and assumptions about those who dropped out being taller or shorter than the mean. Additional secondary analyses will include assessing hazard ratios for events such as mortality and hospitalization, including time to event analysis.

Study timeline
Enrollment began 5 September 2017 and is planned to continue through 5 September 2018. Children will be followed until age 18 months, resulting in a planned completion in February 2020. Blinding will be maintained until all final study visits are completed. The primary analysis will be performed once data collection is complete for the variables involved.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  52  53  54  55  56  57  58  59  60   F  o  r  p  e  e  r  r  e  v  i  e  w  o  n  l  y   24 To reduce intra-assay variability, the majority of laboratory testing will be batched and performed once complete samples are present.

Potential challenges and limitations
Potential challenges include adequate recruitment of target participant numbers within the one-year study window. Should this be true, we will consider extending the recruitment window. Maintaining adherence of study medication, particularly the daily doses of nicotinamide to breast-feeding mothers during months 0-6, and infants during months 6-18, may be a challenge in a setting in which few individuals are accustomed to taking daily medication or nutritional supplements. To protect against adherence challenges, we performed a pilot study beforehand, identifying potential barriers to adherence and resulting in steps to remind mothers frequently during the first month of the study, with follow-up home visits by the field team and by local community healthcare workers.

Patient and Public Involvement
Patients and the public were not directly involved in the planning of this study. Results will be communicated to participants via community meetings with local leaders.

Dissemination
This trial has been registered at ClinicalTrials.gov (#NCT03268902). Scientific reports resulting from this study will be made publicly available. Following completion of data accrual, data will be made publicly available in an online database. Data will be presented at major conferences such as those of the American Society of Tropical Medicine and Hygiene, the Pediatric Academic Society, the Endocrine Society, and the African Academy of Sciences.
338x190mm (300 x 300 DPI)   Role of study sponsor and funders, if any, in study design; collection, management, analysis, and interpretation of data; writing of the report; and the decision to submit the report for publication, including whether they will have ultimate authority over any of these activities ______23______ 5d Composition, roles, and responsibilities of the coordinating centre, steering committee, endpoint adjudication committee, data management team, and other individuals or groups overseeing the trial, if applicable (see Item 21a for data monitoring committee) ______NA______ Description of trial design including type of trial (eg, parallel group, crossover, factorial, single group), allocation ratio, and framework (eg, superiority, equivalence, noninferiority, exploratory) ______7-8_____

Methods: Participants, interventions, and outcomes
Study setting 9 Description of study settings (eg, community clinic, academic hospital) and list of countries where data will be collected. Reference to where list of study sites can be obtained ______9_______ Eligibility criteria 10 Inclusion and exclusion criteria for participants. If applicable, eligibility criteria for study centres and individuals who will perform the interventions (eg, surgeons, psychotherapists)

______8_______
Interventions 11a Interventions for each group with sufficient detail to allow replication, including how and when they will be administered ______9-10____ 11b Criteria for discontinuing or modifying allocated interventions for a given trial participant (eg, drug dose change in response to harms, participant request, or improving/worsening disease)

_____19-20____
11c Strategies to improve adherence to intervention protocols, and any procedures for monitoring adherence (eg, drug tablet return, laboratory tests)

______10______
11d Relevant concomitant care and interventions that are permitted or prohibited during the trial ______10_______ Outcomes 12 Primary, secondary, and other outcomes, including the specific measurement variable (eg, systolic blood pressure), analysis metric (eg, change from baseline, final value, time to event), method of aggregation (eg, median, proportion), and time point for each outcome. Explanation of the clinical relevance of chosen efficacy and harm outcomes is strongly recommended _18-19, Table 1__ Participant timeline 13 Time schedule of enrolment, interventions (including any run-ins and washouts), assessments, and visits for participants. A schematic diagram is highly recommended (see Figure) _____9-10______

Methods: Assignment of interventions (for controlled trials)
Allocation: Sequence generation 16a Method of generating the allocation sequence (eg, computer-generated random numbers), and list of any factors for stratification. To reduce predictability of a random sequence, details of any planned restriction (eg, blocking) should be provided in a separate document that is unavailable to those who enrol participants or assign interventions ______9_______ Allocation concealment mechanism 16b Mechanism of implementing the allocation sequence (eg, central telephone; sequentially numbered, opaque, sealed envelopes), describing any steps to conceal the sequence until interventions are assigned ______9_______ Implementation 16c Who will generate the allocation sequence, who will enrol participants, and who will assign participants to interventions ______9_______ Blinding (masking) 17a Who will be blinded after assignment to interventions (eg, trial participants, care providers, outcome assessors, data analysts), and how ______9-10_____ 17b If blinded, circumstances under which unblinding is permissible, and procedure for revealing a participant's allocated intervention during the trial ______19-20____

Methods: Data collection, management, and analysis
Data collection methods 18a Plans for assessment and collection of outcome, baseline, and other trial data, including any related processes to promote data quality (eg, duplicate measurements, training of assessors) and a description of study instruments (eg, questionnaires, laboratory tests) along with their reliability and validity, if known. Reference to where data collection forms can be found, if not in the protocol ____10-18______ 18b Plans to promote participant retention and complete follow-up, including list of any outcome data to be collected for participants who discontinue or deviate from intervention protocols _____9-10______ Methods for any additional analyses (eg, subgroup and adjusted analyses) _____19_______ 20c Definition of analysis population relating to protocol non-adherence (eg, as randomised analysis), and any statistical methods to handle missing data (eg, multiple imputation) _____19_______

Methods: Monitoring
Data monitoring 21a Composition of data monitoring committee (DMC); summary of its role and reporting structure; statement of whether it is independent from the sponsor and competing interests; and reference to where further details about its charter can be found, if not in the protocol. Alternatively, an explanation of why a DMC is not needed Description of any interim analyses and stopping guidelines, including who will have access to these interim results and make the final decision to terminate the trial _____19-20_____

Harms 22
Plans for collecting, assessing, reporting, and managing solicited and spontaneously reported adverse events and other unintended effects of trial interventions or trial conduct _____19-20_____ Auditing 23 Frequency and procedures for auditing trial conduct, if any, and whether the process will be independent from investigators and the sponsor ______18_______ How personal information about potential and enrolled participants will be collected, shared, and maintained in order to protect confidentiality before, during, and after the trial _____18________ Declaration of interests 28 Financial and other competing interests for principal investigators for the overall trial and each study site _____23________

Ethics and dissemination
Access to data 29 Statement of who will have access to the final trial dataset, and disclosure of contractual agreements that limit such access for investigators _____18________ Ancillary and posttrial care 30 Provisions, if any, for ancillary and post-trial care, and for compensation to those who suffer harm from trial participation

_____19-20_____
Dissemination policy 31a Plans for investigators and sponsor to communicate trial results to participants, healthcare professionals, the public, and other relevant groups (eg, via publication, reporting in results databases, or other data sharing arrangements), including any publication restrictions _______22______ 31b Authorship eligibility guidelines and any intended use of professional writers _______22______ 31c Plans, if any, for granting public access to the full protocol, participant-level dataset, and statistical code _______22______

Appendices
Informed consent materials 32 Model consent form and other related documentation given to participants and authorised surrogates __Supplement__ Biological specimens 33 Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in the current trial and for future use in ancillary studies, if applicable ____10-18______ *It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 Explanation & Elaboration for important clarification on the items. Amendments to the protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT Group under the Creative Commons "Attribution-NonCommercial-NoDerivs 3.0 Unported" license.  1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  23  24  25  26  27  28  29  30  31  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47