phMRI: methodological considerations for mitigating potential confounding factors

Neuroimaging in psychedelic drug development: past, present, and future

Psychedelic therapy (PT) is an emerging paradigm with great transdiagnostic potential for treating psychiatric disorders, including depression, addiction, post-traumatic stress disorder, and potentially others. 'Classic' serotonergic psychedelics, such as psilocybin and lysergic acid diethylamide (LSD), which have a key locus of action at the 5-HT2A receptor, form the main focus of this movement, but substances including ketamine, 3,4-Methylenedioxymethamphetamine (MDMA) and ibogaine also hold promise. The modern phase of development of these treatment modalities in the early 21st century has occurred concurrently with the wider use of advanced human neuroscientific research methods; principally neuroimaging. This can potentially enable assessment of drug and therapy brain effects with greater precision and quantification than any previous novel development in psychiatric pharmacology. We outline the major trends in existing data and suggest the modern development of PT has benefitted greatly from the use of neuroimaging. Important gaps in existing knowledge are identified, namely: the relationship between acute drug effects and longer-term (clinically-relevant) effects, the precise characterisation of effects at the 5-HT2A receptor and relationships with functional/clinical effects, and the possible impact of these compounds on neuroplasticity. A road-map for future research is laid out, outlining clinical studies which will directly address these three questions, principally using combined Positron Emission Tomography (PET) and Magnetic Resonance Imaging (MRI) methods, plus other adjunct techniques. Multimodal (PET/MRI) studies using modern PET techniques such as the 5-HT2A-selective ligand [11 C]Cimbi-36 (and other ligands sensitive to neuroplasticity changes) alongside MRI measures of brain function would provide a 'molecular-functional-clinical bridge' in understanding. Such results would help to resolve some of these questions and provide a firmer foundation for the ongoing development of PT.

Melanocortin 4 receptor agonism enhances sexual brain processing in women with hypoactive sexual desire disorder

BACKGROUND

Hypoactive sexual desire disorder (HSDD) is characterized by a persistent deficiency of sexual fantasies and desire for sexual activity, causing marked distress and interpersonal difficulty. It is the most prevalent female sexual health problem globally, affecting approximately 10% of women, but has limited treatment options. Melanocortin 4 receptor (MC4R) agonists have emerged as a promising therapy for women with HSDD, through unknown mechanisms. Studying the pathways involved is crucial for our understanding of normal and abnormal sexual behavior.

METHODS

Using psychometric, functional neuroimaging, and hormonal analyses, we conducted a randomized, double-blinded, placebo-controlled, crossover clinical study to assess the effects of MC4R agonism compared with placebo on sexual brain processing in 31 premenopausal heterosexual women with HSDD.

RESULTS

MC4R agonism significantly increased sexual desire for up to 24 hours after administration compared with placebo. During functional neuroimaging, MC4R agonism enhanced cerebellar and supplementary motor area activity and deactivated the secondary somatosensory cortex, specifically in response to visual erotic stimuli, compared with placebo. In addition, MC4R agonism enhanced functional connectivity between the amygdala and the insula during visual erotic stimuli compared with placebo.

CONCLUSION

These data suggest that MC4R agonism enhanced sexual brain processing by reducing self-consciousness, increasing sexual imagery, and sensitizing women with HSDD to erotic stimuli. These findings provide mechanistic insight into the action of MC4R agonism in sexual behavior and are relevant to the ongoing development of HSDD therapies and MC4R agonist development more widely.

TRIAL REGISTRATION

ClinicalTrials.gov NCT04179734.

FUNDING

This is an investigator-sponsored study funded by AMAG Pharmaceuticals Inc., the Medical Research Council (MRC) (MR/T006242/1), and the National Institute for Health Research (NIHR) (CS-2018-18-ST2-002 and RP-2014-05-001).

Identification of key somatic oncogenic mutation based on a confounder-free causal inference model

Abnormal cell proliferation and epithelial-mesenchymal transition (EMT) are the essential events that induce cancer initiation and progression. A fundamental goal in cancer research is to develop an efficient method to detect mutational genes capable of driving cancer. Although several computational methods have been proposed to identify these key mutations, many of them focus on the association between genetic mutations and functional changes in relevant biological processes, but not their real causality. Causal effect inference provides a way to estimate the real induce effect of a certain mutation on vital biological processes of cancer initiation and progression, through addressing the confounder bias due to neutral mutations and unobserved latent variables. In this study, integrating genomic and transcriptomic data, we construct a novel causal inference model based on a deep variational autoencoder to identify key oncogenic somatic mutations. Applied to 10 cancer types, our method quantifies the causal effect of genetic mutations on cell proliferation and EMT by reducing both observed and unobserved confounding biases. The experimental results indicate that genes with higher mutation frequency do not necessarily mean they are more potent in inducing cancer and promoting cancer development. Moreover, our study fills a gap in the use of machine learning for causal inference to identify oncogenic mutations.

Identifying key mutations of cancers is helpful to better understand the mechanisms of cancer cell transformation and is critical for therapeutic approaches. Besides sequence and structure-based computational approaches, some functional impact-based methods which consider the association between mutation events and the activity of cancer-related biological processes have also been developed to detect key mutations. However, these methods mainly consider the correlation but ignore that the correlation is far from causality due to the existence of observed and unobserved confounding factors. We develop a confounder-free machine learning-based causal inference framework to estimate the causal effect of mutations on abnormal cell proliferation and epithelial-mesenchymal transition (EMT). It fills a gap in the use of causal mechanisms to discover potential driver mutations in cancer biological systems. Applying our method to 10 cancer types, the identified key mutations are highly consistent with public well-verified ones. Additionally, some new key mutations have also been discovered.

"Less is more": a dose-response account of intranasal oxytocin pharmacodynamics in the human brain

Intranasal oxytocin is attracting attention as a potential treatment for several brain disorders due to promising preclinical results. However, translating findings to humans has been hampered by remaining uncertainties about its pharmacodynamics and the methods used to probe its effects in the human brain. Using a dose-response design (9, 18 and 36 IU), we demonstrate that intranasal oxytocin-induced changes in local regional cerebral blood flow (rCBF) in the amygdala at rest, and in the covariance between rCBF in the amygdala and other key hubs of the brain oxytocin system, follow a dose-response curve with maximal effects for lower doses. Yet, the effects on local rCBF might vary by amygdala subdivision, highlighting the need to qualify dose-response curves within subregion. We further link physiological changes with the density of the oxytocin receptor gene mRNA across brain regions, strengthening our confidence in intranasal oxytocin as a valid approach to engage central targets. Finally, we demonstrate that intranasal oxytocin does not disrupt cerebrovascular reactivity, which corroborates the validity of haemodynamic neuroimaging to probe the effects of intranasal oxytocin in the human brain. DATA AVAILABILITY: Participants did not consent for open sharing of the data. Therefore, data can only be accessed from the corresponding author upon reasonable request.

The effect of risperidone on reward-related brain activity is robust to drug-induced vascular changes

Dopamine (DA) mediated brain activity is intimately linked to reward‐driven cerebral responses, while aberrant reward processing has been implicated in several psychiatric disorders. fMRI has been a valuable tool in understanding the mechanism by which DA modulators alter reward‐driven responses and how they may exert their therapeutic effect. However, the potential effects of a pharmacological compound on aspects of neurovascular coupling may cloud the interpretability of the BOLD contrast. Here, we assess the effects of risperidone on reward driven BOLD signals produced by reward anticipation and outcome, while attempting to control for potential drug effects on regional cerebral blood flow (CBF) and cerebrovascular reactivity (CVR). Healthy male volunteers (n = 21) each received a single oral dose of either 0.5 mg, 2 mg of risperidone or placebo in a double‐blind, placebo‐controlled, randomised, three‐period cross‐over study design. Participants underwent fMRI scanning while performing the widely used Monetary Incentive Delay (MID) task to assess drug impact on reward function. Measures of CBF (Arterial Spin Labelling) and breath‐hold challenge induced BOLD signal changes (as a proxy for CVR) were also acquired and included as covariates. Risperidone produced divergent, dose‐dependent effects on separate phases of reward processing, even after controlling for potential nonneuronal influences on the BOLD signal. These data suggest the D2 antagonist risperidone has a wide‐ranging influence on DA‐mediated reward function independent of nonneuronal factors. We also illustrate that assessment of potential vascular confounds on the BOLD signal may be advantageous when investigating CNS drug action and advocate for the inclusion of these additional measures into future study designs.

Training confounder-free deep learning models for medical applications

The presence of confounding effects (or biases) is one of the most critical challenges in using deep learning to advance discovery in medical imaging studies. Confounders affect the relationship between input data (e.g., brain MRIs) and output variables (e.g., diagnosis). Improper modeling of those relationships often results in spurious and biased associations. Traditional machine learning and statistical models minimize the impact of confounders by, for example, matching data sets, stratifying data, or residualizing imaging measurements. Alternative strategies are needed for state-of-the-art deep learning models that use end-to-end training to automatically extract informative features from large set of images. In this article, we introduce an end-to-end approach for deriving features invariant to confounding factors while accounting for intrinsic correlations between the confounder(s) and prediction outcome. The method does so by exploiting concepts from traditional statistical methods and recent fair machine learning schemes. We evaluate the method on predicting the diagnosis of HIV solely from Magnetic Resonance Images (MRIs), identifying morphological sex differences in adolescence from those of the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA), and determining the bone age from X-ray images of children. The results show that our method can accurately predict while reducing biases associated with confounders. The code is available at https://github.com/qingyuzhao/br-net .

Pharmaco-fMRI: A Tool to Predict the Response to Antiepileptic Drugs in Epilepsy

Pharmacological treatment with antiepileptic medications (AEDs) in epilepsy is associated with a variety of neurocognitive side effects. However, the mechanisms underlying these side effects, and why certain brain anatomies are more affected still remain poorly understood. Advanced functional magnetic resonance imaging (fMRI) methods, such as pharmaco-fMRI, can investigate medication-related effects on brain activities using task and resting state fMRI and showing reproducible activation and deactivation patterns. This methodological approach has been used successfully to complement neuropsychological studies of AEDs. Here we review pharmaco-fMRI studies in people with epilepsy targeting the most-widely prescribed AEDs. Pharmco-fMRI has advanced our understanding of the impact of AEDs on specific brain networks and thus may provide potential biomarkers to move beyond the current “trial and error” approach when commencing anti-epileptic medication.

Role of brain imaging in disorders of brain-gut interaction: a Rome Working Team Report

Imaging of the living human brain is a powerful tool to probe the interactions between brain, gut and microbiome in health and in disorders of brain-gut interactions, in particular IBS. While altered signals from the viscera contribute to clinical symptoms, the brain integrates these interoceptive signals with emotional, cognitive and memory related inputs in a non-linear fashion to produce symptoms. Tremendous progress has occurred in the development of new imaging techniques that look at structural, functional and metabolic properties of brain regions and networks. Standardisation in image acquisition and advances in computational approaches has made it possible to study large data sets of imaging studies, identify network properties and integrate them with non-imaging data. These approaches are beginning to generate brain signatures in IBS that share some features with those obtained in other often overlapping chronic pain disorders such as urological pelvic pain syndromes and vulvodynia, suggesting shared mechanisms. Despite this progress, the identification of preclinical vulnerability factors and outcome predictors has been slow. To overcome current obstacles, the creation of consortia and the generation of standardised multisite repositories for brain imaging and metadata from multisite studies are required.

Dissociable effects of cannabis with and without cannabidiol on the human brain's resting-state functional connectivity

BACKGROUND

Two major constituents of cannabis are Δ⁹-tetrahydrocannabinol (THC) and cannabidiol (CBD). THC is the main psychoactive component; CBD may buffer the user against the harmful effects of THC.

AIMS

We examined the effects of two strains of cannabis and placebo on the human brain's resting-state networks using fMRI.

METHODS

Seventeen healthy volunteers (experienced with cannabis, but not regular users) underwent three drug treatments and scanning sessions. Treatments were cannabis containing THC (Cann-CBD; 8 mg THC), cannabis containing THC with CBD (Cann+CBD; 8 mg THC + 10 mg CBD), and matched placebo cannabis. Seed-based resting-state functional connectivity analyses were performed on three brain networks: the default mode (DMN; defined by positive connectivity with the posterior cingulate cortex: PCC+), executive control (ECN; defined by negative connectivity with the posterior cingulate cortex: PCC-) and salience (SAL; defined by positive connectivity with the anterior insula: AI+) network.

RESULTS

Reductions in functional connectivity (relative to placebo) were seen in the DMN (PCC+) and SAL (AI+) networks for both strains of cannabis, with spatially dissociable effects. Across the entire salience network (AI+), Cann-CBD reduced connectivity relative to Cann+CBD. The PCC in the DMN was specifically disrupted by Cann-CBD, and this effect correlated with subjective drug effects, including feeling 'stoned' and 'high'.

CONCLUSIONS

THC disrupts the DMN, and the PCC is a key brain region involved in the subjective experience of THC intoxication. CBD restores disruption of the salience network by THC, which may explain its potential to treat disorders of salience such as psychosis and addiction.

Effects of high-dose baclofen on cue reactivity in alcohol dependence: A randomized, placebo-controlled pharmaco-fMRI study

Increased functional brain response towards alcohol-associated stimuli is a neural hallmark of alcohol dependence and a promising target for pharmacotherapy. For the first time, we assessed the effects of individually titrated high-dose baclofen on cue reactivity and functional connectivity in alcohol-dependent (AD) patients in a randomized controlled trial (RCT). We investigated 23 recently detoxified AD patients and 23 matched healthy controls (HC) with a cue reactivity functional magnetic resonance imaging task. Patients were further scanned at baseline without medication and during treatment with high-dose baclofen/placebo (30-270 mg/d). Analyses were conducted for alcohol cue-elicited brain response, alcohol cue-modulated and stimulus-independent functional connectivity with left ventral tegmental area (VTA) as seed region. At baseline, AD patients (N = 23) showed increased cue-elicited brain activation in the ventral striatum (VS) compared to HC (N = 23), which was decreased at the second scanning session compared to baseline. Patients receiving baclofen (N = 10) showed a significant stronger decrease in cue-elicited brain activation in left orbitofrontal cortex (OFC), bilateral amygdala and left VTA than patients receiving placebo (N = 13). Treatment with baclofen further led to a decrease in alcohol cue-modulated functional connectivity between left VTA and left anterior cingulate cortex (ACC) as well as left medial prefrontal cortex (MPFC). Regarding clinical outcome, significantly more patients of the baclofen group remained abstinent during the high-dose period. Baclofen specifically decreased cue-elicited brain responses in areas known to be involved in the processing of salient (appetitive and aversive) stimuli. Treatment with high-dose baclofen seems to provide a pharmacological relief of this neural "warning signal" evoked by alcohol-related cues, thereby possibly supporting patients in remaining abstinent. Trial Registration Identifier of the main trial [BACLAD study] at clinicaltrials.gov: NCT01266655.

A short, robust brain activation control task optimised for pharmacological fMRI studies

BACKGROUND

Functional magnetic resonance imaging (fMRI) is a popular method for examining pharmacological effects on the brain; however, the BOLD response is dependent on intact neurovascular coupling, and potentially modulated by a number of physiological factors. Pharmacological fMRI is therefore vulnerable to confounding effects of pharmacological probes on general physiology or neurovascular coupling. Controlling for such non-specific effects in pharmacological fMRI studies is therefore an important consideration, and there is an additional need for well-validated fMRI task paradigms that could be used to control for such effects, or for general testing purposes.

METHODS

We have developed two variants of a standardized control task that are short (5 minutes duration) simple (for both the subject and experimenter), widely applicable, and yield a number of readouts in a spatially diverse set of brain networks. The tasks consist of four functionally discrete three-second trial types (plus additional null trials) and contain visual, auditory, motor and cognitive (eye-movements, and working memory tasks in the two task variants) stimuli. Performance of the tasks was assessed in a group of 15 subjects scanned on two separate occasions, with test-retest reliability explicitly assessed using intra-class correlation coefficients.

RESULTS

Both tasks produced robust patterns of brain activation in the expected brain regions, and region of interest-derived reliability coefficients for the tasks were generally high, with four out of eight task conditions rated as 'excellent' or 'good', and only one out of eight rated as 'poor'. Median values in the voxel-wise reliability measures were also >0.7 for all task conditions, and therefore classed as 'excellent' or 'good'. The spatial concordance between the most highly activated voxels and those with the highest reliability coefficients was greater for the sensory (auditory, visual) conditions than the other (motor, cognitive) conditions.

DISCUSSION

Either of the two task variants would be suitable for use as a control task in future pharmacological fMRI studies or for any other investigation where a short, reliable, basic task paradigm is required. Stimulus code is available online for re-use by the scientific community.

Functional Characterization of 5-HT1B Receptor Drugs in Nonhuman Primates Using Simultaneous PET-MR

In the present study, we used a simultaneous PET-MR experimental design to investigate the effects of functionally different compounds (agonist, partial agonist, and antagonist) on 5-HT_1B receptor (5-HT_1BR) occupancy and the associated hemodynamic responses. In anesthetized male nonhuman primates (n = 3), we used positron emission tomography (PET) imaging with the radioligand [¹¹C]AZ10419369 administered as a bolus followed by constant infusion to measure changes in 5-HT_1BR occupancy. Simultaneously, we measured changes in cerebral blood volume (CBV) as a proxy of drug effects on neuronal activity. The 5-HT_1BR partial agonist AZ10419369 elicited a dose-dependent biphasic hemodynamic response that was related to the 5-HT_1BR occupancy. The magnitude of the response was spatially overlapping with high cerebral 5-HT_1BR densities. High doses of AZ10419369 exerted an extracranial tissue vasoconstriction that was comparable to the less blood-brain barrier-permeable 5-HT_1BR agonist sumatriptan. By contrast, injection of the antagonist GR127935 did not elicit significant hemodynamic responses, even at a 5-HT_1BR cerebral occupancy similar to the one obtained with a high dose of AZ10419369. Given the knowledge we have of the 5-HT_1BR and its function and distribution in the brain, the hemodynamic response informs us about the functionality of the given drug: changes in CBV are only produced when the receptor is stimulated by the partial agonist AZ10419369 and not by the antagonist GR127935, consistent with low basal occupancy by endogenous serotonin.SIGNIFICANCE STATEMENT We here show that combined simultaneous positron emission tomography and magnetic resonance imaging uniquely enables the assessment of CNS active compounds. We conducted a series of pharmacological interventions to interrogate 5-HT_1B receptor binding and function and determined blood-brain barrier passage of drugs and demonstrate target involvement. Importantly, we show how the spatial and temporal effects on brain hemodynamics provide information about pharmacologically driven downstream CNS drug effects; the brain hemodynamic response shows characteristic dose-related effects that differ depending on agonistic or antagonistic drug characteristics and on local 5-HT_1B receptor density. The technique lends itself to a comprehensive in vivo investigation and understanding of drugs' effects in the brain.

Intrinsic functional connectivity in late-life depression: trajectories over the course of pharmacotherapy in remitters and non-remitters

Previous studies in late-life depression (LLD) have found that patients have altered intrinsic functional connectivity in the dorsal default mode network (DMN) and executive control network (ECN). We aimed to detect connectivity differences across a treatment trial among LLD patients as a function of remission status. LLD patients (N=37) were enrolled into a 12-week trial of venlafaxine and underwent five functional magnetic resonance imaging resting state scans during treatment. Patients had no history of drug abuse, psychosis, dementia/neurodegenerative diseases or medical conditions with known effects on mood. We investigated whether there were differences in three networks: DMN, ECN and anterior salience network connectivity, as well as a whole brain centrality measure (eigenvector centrality). We found that remitters showed increases in ECN connectivity in the right precentral gyrus and decreases in DMN connectivity in the right inferior frontal gyrus and supramarginal gyrus. The ECN and DMN had regions (middle temporal gyrus and bilateral middle/inferior temporal/fusiform gyrus, respectively) that showed reversed effects (decreased ECN and increased DMN, respectively). Early changes in functional connectivity can occur after initial medication exposure. This study offers new data, indicating that functional connectivity changes differ depending on treatment response and can occur shortly after exposure to antidepressant medication.

How do antidepressants influence the BOLD signal in the developing brain?

Depression is a highly prevalent life-threatening disorder, with its first onset commonly occurring during adolescence. Adolescent depression is increasingly being treated with antidepressants, such as fluoxetine. The use of medication during this sensitive period of physiological and cognitive brain development produces neurobiological changes, some of which may outlast the course of treatment. In this review, we look at how antidepressant treatment in adolescence is likely to alter neurovascular coupling and brain energy use and how these changes, in turn, affect our ability to identify neuronal activity changes between participant groups. BOLD (blood oxygen level dependent) fMRI (functional magnetic resonance imaging), the method most commonly used to record brain activity in humans, is an indirect measure of neuronal activity. This means that between-group comparisons – adolescent versus adult, depressed versus healthy, medicated versus non-medicated – rely upon a stable relationship existing between neuronal activity and the BOLD response across these groups. We use data from animal studies to detail the ways in which fluoxetine may alter this relationship, and explore how these alterations may influence the interpretation of BOLD signal differences between groups that have been treated with fluoxetine and those that have not.

How the heart speaks to the brain: neural activity during cardiorespiratory interoceptive stimulation

Prominent theories emphasize key roles for the insular cortex in the central representation of interoceptive sensations, but how this brain region responds dynamically to changes in interoceptive state remains incompletely understood. Here, we systematically modulated cardiorespiratory sensations in humans using bolus infusions of isoproterenol, a rapidly acting peripheral beta-adrenergic agonist similar to adrenaline. To identify central neural processes underlying these parametrically modulated interoceptive states, we used pharmacological functional magnetic resonance imaging (phMRI) to simultaneously measure blood-oxygenation-level dependent (BOLD) and arterial spin labelling (ASL) signals in healthy participants. Isoproterenol infusions induced dose-dependent increases in heart rate and cardiorespiratory interoception, with all participants endorsing increased sensations at the highest dose. These reports were accompanied by increased BOLD and ASL activation of the right insular cortex at the highest dose. Different responses across insula subregions were also observed. During anticipation, insula activation increased in more anterior regions. During stimulation, activation increased in the mid-dorsal and posterior insula on the right, but decreased in the same regions on the left. This study demonstrates the feasibility of phMRI for assessing brain activation during adrenergic interoceptive stimulation, and provides further evidence supporting a dynamic role for the insula in representing changes in cardiorespiratory states.This article is part of the themed issue 'Interoception beyond homeostasis: affect, cognition and mental health'.