Safety and target engagement of an oral small-molecule sequestrant in adolescents with autism spectrum disorder: an open-label phase 1b/2a trial | Nature Medicine
Mục Lục
Preclinical methods
Mouse husbandry
All animal husbandry and experiments were approved by the California Institute of Technology Institutional Animal Care and Use Committee. Throughout the study, colonized animals were maintained in autoclaved microisolator cages with autoclaved bedding (Aspen Chip Bedding, Northeastern Products), water and chow. Standard chow was provided to the animals (Laboratory Autoclavable Rodent Diet 5010, LabDiet) until diet switch to irradiated 5% AB-2004 or control diets (Teklad) at 5 weeks. This percentage of AB-2004 (AST-120) in mouse chow has been previously used safely in mice34,86,87. Mice were maintained at an ambient temperature of 71–75 °F, 30–70% humidity, at a cycle of 13 h light and 11 h dark.
Experimental design of mouse experiments
Germ-free C57BL/6J male mouse (Mus musculus) weanlings (3 weeks of age) from the Mazmanian laboratory colony were colonized by gavage of 100 µl of a 1:1 mixture of 109 colony-forming units per milliliter of Bacteroides ovatus (± 4EP pathway genes) and wild-type Lactiplantibacillus plantarum. Urine was collected at 7 weeks before behavior testing. Behavior testing began at 7 weeks of age, 3 d after urine collection.
Analysis of metabolites from urine of mice
Urine was passively collected, and 4EPS levels were quantified by liquid chromatography–mass spectrometry (LC–MS) and normalized to creatinine levels by Charles River Laboratories.
Behavior testing
Behavior testing was performed as previously described34,86,87. All mice were tested using the same battery of behavioral tests, starting at 6 weeks of age, in the following order: EPM, open-field testing, marble burying and grooming. Mice were allowed to settle for at least 2 d after cage changing before they were tested, and tests were performed 2–3 d apart to allow mice to rest between tests. Mice were acclimated to the behavior testing room for 1 h before testing. Mice were tested during the light phase of the light/dark cycle.
Clinical methods
Clinical study design and ethical approval
The AXL-2004-001 study (trial registration no. ACTRN12618001956291) (ANZCTR (88,89)) was an open-label, outpatient, multiple-ascending-dose phase 1b/2a study in an ASD-diagnosed adolescent (12–17 years old) population with confirmed GI symptoms (for example, diarrhea, constipation, abdominal pain and bloating). Forty-one individuals were screened between 18 April 2019 and 23 January 2020. Thirty participants were enrolled across three sites in Australia and New Zealand, including the Queensland Children’s Hospital in Brisbane (14 participants), the Brain and Mind Centre in Sydney (six participants) and Optimal Clinical Trials in Auckland (ten participants). There was no formal sample size calculation for this phase 1 study because it focused on safety and tolerability. This approach is common in early-stage exploratory clinical trials. All necessary licenses and permissions to use the behavioral assessments outlined in the study protocol were obtained before initiating the study.
The study protocol, investigator brochure, participant information and consent forms, participant-facing questionnaires, recruitment documentation and procedures and documentation regarding the investigatorsʼ experience and qualifications were submitted to Health and Disability Ethics Committees (New Zealand), Children’s Health Queensland Hospital and the Health Service Human Research Ethics Committee and Bellberry Human Research Ethics Committee for ethical review and approval. The study was conducted in accordance with the Declaration of Helsinki (Fortaleza, October 2013), ICH E6 guidelines, Good Clinical Practice and local regulations.
Study participation
This open-label study consisted of four different dosing plans based on participant weight at visit 1. Eligible participants were escalated through three dosing periods during the 8-week treatment period, starting with the lowest dose for their dosing plan (see Supplementary Methods for more details). Participants were requested to consume AB-2004 90 min after any other concomitant medications. Safety and tolerability were confirmed before a participant escalated to the next dosing level. If participants were unable to tolerate a dosing level, they were returned to the previous dosing level for the remainder of the treatment period. After the last dose of AB-2004, participants returned to the clinic 28 d later for a follow-up safety evaluation (FV). The last visit of the study was completed on 15 May 2020. Patient data were collected using IMednet (version 1.94.0). An e-diary by Dedo (www.dedo.ai) was designed to collect GI data.
Study participants and study populations
A total of 41 adolescent individuals, aged 12–17 years inclusive, were screened for eligibility for participation in the study, and the 30 who met the study-specific eligibility criteria were enrolled and received at least one dose of AB-2004 (Safety Population). Of the 41 individuals screened and 30 enrolled, 40 and 29, respectively, were male. A predominantly male cohort was targeted to reduce variability in response in this exploratory study that surveyed a wide range of behavioral assessments. One participant withdrew after the first dose due to the investigator’s decision based on the participant presenting with an unrelated viral infection. Another participant withdrew consent during the low-dose period due to anticipated admission to hospital for pre-existing behavioral difficulties. One participant withdrew due to significant study non-compliance, and two did not complete FV assessments due to the caregivers being unwell and unable to accompany the participants. A total of 27 participants (26 males and one female) completed at least up to the EOT visit (Completers Population). One participant, the female participant, was included in the Safety Population but was not included in the exploratory efficacy analysis. This participant was removed from the exploratory efficacy analysis because their participation in the trial coincided with the initial COVID-19 pandemic outbreak and its associated societal restrictions put into effect in Australia. These restrictions prevented the participant from conducting normal routines and accessing normal services. As determined by the site principal investigator, these abrupt changes in routine had an effect on the behavior of the participant; therefore, this participant was excluded from the efficacy analysis.
Safety assessments
The primary endpoint of the study was the safety and tolerability of AB-2004 as assessed by physical examinations, vital signs, clinical laboratory measurements (hematology, serum chemistry and urinalysis) and adverse events.
Blood collection
Blood was obtained using uniform collection kits from Sonic Clinical Trials sent to each facility. Blood was drawn from study participants on visits 1, 4 and 5 and aliquoted for health monitoring by Sonic Clinical Trials and metabolite analysis by Metabolon. Blood chemistry panels performed by Sonic Clinical Trials included albumin, alkaline phosphatase, alanine amino transferase, aspartate amino transferase, blood urea nitrogen, urea, corrected calcium, bicarbonate, chloride, creatinine, gamma-glutamyl transpeptidase, glucose, lactate dehydrogenase, magnesium, phosphorus, potassium, sodium, total bilirubin, conjugated bilirubin, unconjugated bilirubin and total protein. Hematology panels included measurement of platelets, hematocrit, red blood cells, hemoglobin, reticulocytes, total white blood cell count and absolute and percentages of neutrophils, lymphocytes, monocytes, eosinophils and basophils.
Urine collection
Participants were provided with a urine home collection kit and instructions to collect all of the first morning void a maximum of 2 d before clinic visit and place in a refrigerator to bring to their visit or to be picked up by courier. Urinalysis samples were collected during the in-clinic visit. Aliquoting for metabolite analysis and health monitoring urinalysis was performed by Sonic Clinical Trials and included measurements of pH, specific gravity, ketones, protein, glucose, nitrite, urobilinogen, leukocyte esterase and blood.
Human plasma metabolite quantification
Human plasma was analyzed by Metabolon. In brief, plasma was spiked with internal standards (4-ethylphenyl sulfate-d4, p-cresol sulfate-d7, 3-hydroxyhippurate-13C2,15N, 3-hydroxyphenylacetate-d3, 3-(3-hydroxyphenyl)-3-hydroxypropionate-d3, 3-indoxyl sulfate-13C6, 3-(4-hydroxyphenyl)propionate-d4, p-cresol glucuronide-d7 and N-acetylserine-d3,), protein precipitated and analyzed on an Agilent 1290/AB Sciex 5500 QTrap LC–MS/MS system equipped with a UHPLC C18 column. Quantitation was performed using a weighted linear least squares regression analysis with a weighting of 1/x or 1/x2 generated from fortified calibration standards prepared immediately before each run.
Human urine metabolite quantification
Human urine was analyzed by Metabolon. In brief, urine was diluted ten-fold and spiked with internal standards (p-cresol sulfate-d7, 3-hydroxyhippurate-13C2,15N, 3-hydroxyphenylacetate-d3, 3-(3-hydroxyphenyl)-3-hydroxypropionate-d3, 3-indoxyl sulfate-13C6, 3-(4-hydroxyphenyl)propionate-d4, p-cresol glucuronide-d7 and N-acetylserine-d3,), and then an aliquot was subjected to either a solvent crash (for p-cresol sulfate, 3-indoxyl sulfate and p-cresol glucuronide) or derivatization (for 3-hydroxyhippurate, 3-hydroxyphenylacetate, 3-(3-hydroxyphenyl)-3-hydroxypropionate, N-acetylserine and 3-(4-hydroxyphenyl)propionate) and analyzed on an Agilent 1290/AB Sciex 5500 QTrap LC–MS/MS system equipped with a UHPLC C18 column in negative mode. Quantification of 4EPS was performed by the same method with a solvent crash (using the internal standard, 4-ethylphenyl sulfate-d4) but without sample dilution. Quantitation was performed using a weighted linear least squares regression analysis with a weighting of 1/x generated from fortified calibration standards prepared immediately before each run. All urine metabolites were normalized to creatinine levels.
Exploratory efficacy assessments
Exploratory efficacy outcomes included changes from BL at EOT and FV on the GSI-6, Numerical Rating Scale (NRS), GSRS, BSS, RBS-R VABS, CASI-5, SRS, CGI-S and CGI-I, ABC or PARS diagnostics. Efficacy assessments were administered on site at the respective clinics during visits. VABS, PARS and CGI-S and CGI-I were conducted by the principal investigator or qualified designee. The GSI-6, NRS, GSRS, RBS-R, BSS, CASI-5 SRS and ABC questionnaires were completed by the designated caregivers of the participants. In the VABS assessment, ten participants did not pass the under 25% estimated answers criterion of any domain during assessment and, thus, had to be removed from this analysis, according to the VABS manual, page 47 (ref. 5).
AB-2004 treatment dosage
For individuals weighing ≥60 kg, three daily doses each of:
Period 1: 0.75 g, Days 1–14 (2 weeks)
Period 2: 1.5 g, Days 15–28 (2 weeks)
Period 3: 2 g, Days 29–56 (4 weeks)
For individuals weighing 50–59 kg, three daily doses each of:
Period 1: 0.75 g, Days 1–14 (2 weeks)
Period 2: 1.0 g, Days 15–28 (2 weeks)
Period 3: 1.75 g, Days 29–56 (4 weeks)
For individuals weighing 40–49 kg, three daily doses each of:
Period 1: 0.5 g, Days 1–14 (2 weeks)
Period 2: 0.75 g, Days 15–28 (2 weeks)
Period 3: 1.5 g, Days 29–56 (4 weeks)
For individuals weighing 30–39 kg, three daily doses each of:
Period 1: 0.5 g, Days 1–14 (2 weeks)
Period 2: 0.75 g, Days 15–28 (2 weeks)
Period 3: 1.0 g, Days 29–56 (4 weeks)
Magnetic resonance
Scan parameters
All scans were collected using phased array receive-only head coils (32 channels at sites 1 and 2, 64 channels at site 3).
High-resolution anatomic images (T1-weighted (T1w) and T2-weighted (T2w)) were acquired with 1-mm isotropic resolution. T1w images (2@4:01 each, for sites 1 and 2, and 1@4:00 for site 3) were sagittally oriented using a 3D MPRAGE sequence. A single resolution matched T2w image (4:28) was acquired (the T2_CUBE sequence at site 1, T2_SPACE at sites 2 and 3).
Two gradient echo multi-band echo planar imaging (EPI) rs-FMRI acquisitions (300 volumes each) were performed with 2.5-mm isotropic resolution, 1-s repetition time and multi-band factor 3. In total, 51 slices were acquired obliquely, with the bottom slice oriented on the line between the bottom of the cerebellum and the bottom of the orbitofrontal cortex. The phase encode was reversed between the first and second scan (AP for the first scan, PA for the second) to allow for distortion correction.
Two diffusion scans were also acquired as part of the protocol (5:56 each), but they were not used for this analysis.
Data processing
Before processing, all data were named and organized following the BIDS 1.2.1 specification. Anatomical and fMRI data used in this manuscript were pre-processed using fMRIPrep 20.0.4 (ref. 90) (RRID: SCR_016216), which is based on Nipype 1.4.2 (ref. 91) (RRID: SCR_002502).
Anatomical data pre-processing
A total of two T1w images were corrected for intensity non-uniformity (INU) with N4BiasFieldCorrection92, distributed with ANTs 2.2.0 (ref. 93) (RRID: SCR_004757) per individual. The T1w reference was then skull-stripped with a Nipype implementation of the antsBrainExtraction.shworkflow (from ANTs), using OASIS30ANTs as target template. Brain tissue segmentation of cerebrospinal fluid (CSF), white matter (WM) and gray matter (GM) was performed on the brain-extracted T1w using fast (FSL 5.0.9, RRID: SCR_002823)94. A T1w reference map was computed after registration of four T1w images (after INU correction) using mri_robust_template (FreeSurfer 6.0.1)95. Brain surfaces were reconstructed using recon-all (FreeSurfer 6.0.1, RRID: SCR_001847)96, and the brain mask estimated previously was refined with a custom variation of the method to reconcile ANTs-derived and FreeSurfer-derived segmentations of the cortical GM of Mindboggle (RRID: SCR_002438)97. Volume-based spatial normalization to two standard spaces (MNI152NLin6Asym and MNI152NLin2009cAsym) was performed through non-linear registration with antsRegistration (ANTs 2.2.0), using brain-extracted versions of both T1w reference and the T1w template. The following templates were selected for spatial normalization: FSL’s MNI ICBM 152 non-linear 6th Generation Asymmetric Average Brain Stereotaxic Registration Model98 (RRID: SCR_002823; TemplateFlow ID: MNI152NLin6Asym) and ICBM 152 Nonlinear Asymmetrical template version 2009c98,99 (RRID:SCR_008796; TemplateFlow ID: MNI152NLin2009cAsym).
Functional data pre-processing
For each of the four blood oxygenation level dependent (BOLD) runs found per participant (across all tasks and sessions), a reference volume and its skull-stripped version were generated using fMRIPrep. A deformation field to correct for susceptibility distortions was estimated based on fMRIPrep’s fieldmap-less approach. Registration is performed with antsRegistration (ANTs 2.2.0), and the process is regularized by constraining deformation to be non-zero only along the phase-encoding direction and modulated with an average fieldmap template100. A corrected EPI reference was calculated. The BOLD reference was then co-registered to the T1w reference using bbregister (FreeSurfer)101. Co-registration was configured with 6 degrees of freedom. Head motion parameters with respect to the BOLD reference were estimated before any spatiotemporal filtering using mcflirt (FSL 5.0.9)102. BOLD runs were slice time corrected using 3dTshift from AFNI 20160207 (ref. 103) (RRID: SCR_005927). The BOLD time series were resampled onto their original, native space by applying a single composite transform to correct for head motion and susceptibility distortions. The BOLD time series were resampled into standard space, generating a pre-processed BOLD run in MNI152NLin6Asym space. First, a reference volume and its skull-stripped version were generated using a custom methodology of fMRIPrep. Automatic removal of motion artifacts using independent component analysis (ICA-AROMA)104 was performed on the pre-processed BOLD on MNI space time series after removal of non-steady-state volumes and spatial smoothing with an isotropic, Gaussian kernel of 6 mm full width at half maximum. Corresponding ‘non-aggressively’ de-noised runs were produced after such smoothing. Additionally, the ‘aggressive’ noise regressors were collected and placed in the corresponding confounds file. Several confounding time series were calculated based on the pre-processed BOLD: framewise displacement (FD), DVARS and three region-wise global signals. FD and DVARS were calculated for each functional run, both using their implementations in Nipype (following the definitions by Power et al.105). The three global signals were extracted within the CSF, the WM and the whole brain masks. Additionally, a set of physiological regressors was extracted to allow for component-based noise correction (CompCor)106. Principal components were estimated after high-pass filtering the pre-processed BOLD time series (using a discrete cosine filter with 128-s cutoff) for the two CompCor variants: temporal (tCompCor) and anatomical (aCompCor). tCompCor components were then calculated from the top 5% variable voxels within a mask covering the subcortical regions. This subcortical mask was obtained by heavily eroding the brain mask. For aCompCor, components were calculated within the intersection of the mask and the union of CSF and WM masks calculated in T1w space, after their projection to the native space of each functional run (using the inverse BOLD-to-T1w transformation). Components were also calculated separately within the WM and CSF masks. For each CompCor decomposition, the k components with the largest singular values were retained, such that the retained components’ time series were sufficient to explain 50% of variance across the nuisance mask (CSF, WM, combined or temporal). The head motion estimates calculated in the correction step were also placed within the corresponding confounds file. The confound time series derived from head motion estimates and global signals were expanded with the inclusion of temporal derivatives and quadratic terms for each107. Frames that exceeded a threshold of 0.5 mm FD or 1.5 standardized DVARS were annotated as motion outliers. Gridded (volumetric) resamplings were performed using antsApplyTransforms (ANTs) and configured with Lanczos interpolation to minimize the smoothing effects of other kernels107. Non-gridded (surface) resamplings were performed using mri_vol2surf (FreeSurfer).
fMRI data analysis
To quantify connectivity between the bilateral amygdala and rACC, a region of interest (ROI) approach was used employing methods from our previous work108. The bilateral amygdala was defined using the Harvard-Oxford atlas. The rACC ROI is just anterior to the genu of the corpus callosum and was used in our previous work108,109. Average time courses for each ROI were extracted, demeaned, detrended, Hamming windowed and correlated to generate a single correlation value (r) for each participant both before and after treatment.
Statistical information
Results presented here are from post hoc analyses of the data from the clinical trial using GraphPad Prism 9. Here we present bar graphs representing the preclinical data by mean ± s.e.m. analyzed by ordinary two-way ANOVA test with false discovery rate correction using the Benjamini, Krieger and Yekutieli method, with individual variances computed for each comparison. Clinical data are presented as mean, and 95% confidence intervals were analyzed by repeated-measures ANOVA or linear mixed-effects model, with Geisser–Greenhouse correction tests and false discovery rate correction by the Benjamini, Krieger and Yekutieli method. Metabolite data are presented as individual graphs but were statistically analyzed across all metabolites and samples. Clinical behavioral metrics were analyzed within each test. Two-tailed Pearson’s correlations were performed comparing change in metabolite levels to change in behavioral scores for the PARS and ABC-I tests. fMRI values were analyzed using a two-tailed paired t-test. Study participants were studied as a single group, and all comparisons, especially those within the subgroup of participants in the top quartile of ASD severity, were post hoc and exploratory in nature. Missing data were not imputed, and data were analyzed for individuals who withdrew from the study, for any reason before study completion, regardless of treatment duration, up to the point of discontinuation.
Reporting Summary
Further information on research design is available in the Nature Research Reporting Summary linked to this article.
