Comments and ControversiesPET Neuroimaging: The White Elephant Packs His Trunk?
Introduction
NeuroImage has for two decades been a favored site for the publication of innovative findings in molecular imaging, although this contribution was always numerically small in comparison to MR-based imaging studies; in recent years only some 5% of NeuroImage articles principally involve applications or refinements of positron emission tomography (PET), along with a very small number of reports on single photon emission computed tomography (SPECT). Faced with the burgeoning submissions of reports on functional and anatomic MR, the publishers and editorial board of NeuroImage have recently launched the open-access journal Neuroimage: Clinical, with the intention of reserving NeuroImage proper for studies of structure-function and brain-behavior relationships. The main criterion for publication in NeuroImage is advancement of our understanding of normal brain function, organization, and structure, but NeuroImage also welcomes papers that explicitly address these questions in animal models of human disease. Papers concerning mainly neuropathology, abnormal development, use of biomarkers for studying disease progression or other matters of direct clinical significance should now be referred to NeuroImage: Clinical.
The NeuroImage editorial board has noted a paucity of PET submissions within the present mandate, and has raised some reservations about the overall importance of PET in advancing knowledge of normal brain function. This concern gives one pause to reflect upon the past contributions of molecular imaging, its current applications, and the future role to be anticipated for PET in studying normal brain function. Since the inception of NeuroImage, MR technology has developed at a dizzying pace, and a variety of functional as well as structural imaging and analysis techniques are now widely available. Meanwhile, PET technology has attained maturity with the adoption of consensus nomenclature for kinetic terms (Innis et al., 2007), and the advent of improved methods for image reconstruction, and kinetic analysis (Lucignani, 2006). However, the strict interpretation of GMP regulations arguably unsuited for microdosing studies may have become an impediment to basic PET research (Långström and Hartvig, 2008), and the great expense and technical difficulty in carrying out PET studies is well-known. Indeed, cost-benefit considerations cannot be ignored in the design of brain PET research, and may likely merit a formal economic analysis, which would entail developing a metric for scientific value gained as a function of cost per study.
While such an economic analysis remains to be developed, it is self-evident that the expense, technical difficulty and radiation safety issues arising from PET are justified when no other method will serve to answer worthwhile questions. The issue at hand then becomes a matter of identifying the purposes to which PET is uniquely suited in studies of normal brain function, thus falling within the constrained purview of NeuroImage, or furnishes insights into human disease, which is the theme of Neuroimage: Clinical. To this end, I have attempted a dispassionate case-by-case analysis of articles recently appearing in the pages of this journal. In particular, I identified the 78 NeuroImage articles appearing online or in print during 2012 for which the acronym PET appeared in the abstract, keywords or title. I then categorized these papers based on my rough assessment of their main theme; in eight of the 78 cases, the articles seemed to fall cleanly between two of my ad hoc criteria, and are thus reported twice. In addition to this thematic analysis of NeuroImage in 2012, I also made a historical survey of the penetration of PET in the greater scientific literature, based on a simple PubMed keyword search. Some patterns emerging from this exercise may guide and inform the trajectory of publishing brain PET research in NeuroImage and elsewhere.
Section snippets
General PET reviews
Two review articles focusing on PET appeared in NeuroImage during 2012, both composed by illustrious researchers, and commissioned for a 20th anniversary volume. Brain oxygen metabolism is the topic of one invited PET review (Baron and Jones, 2012). Kinetic studies of inhaled [15O]-oxygen afford calculation of the cerebral metabolic rate (CMRO2), and in conjunction with cerebral blood flow (CBF) measurements and also the oxygen extraction fraction (OEF) in tissues. Early success with these
Incidental mentions of PET
In 16 of the 78 (21%) NeuroImage articles from 2012 mentioning PET in the abstract, this mention is only incidental, in providing the background or motivation of the new study. Overall, these reports bear witness to the great methodological debt of fMRI to the solutions developed for comparable problems in PET research. In one review article, the PET studies of flow-metabolism coupling cited above are specifically presented as the inspiration of fMRI measurements of BOLD signal changes (Fox,
New PET techniques and tracers
In 29 of the 78 (37%) cases from the sample, the main topic concerned new PET techniques, such as the development of novel PET ligands and kinetic modeling procedures (N = 11), the endogenous ligand competition paradigm noted above, along with occupancy studies (N = 7), input modeling (N = 3) and PET instrumentation (N = 6). The first category included a trio of reports on new tracers for opioid receptor like-1 in non-human primates and in humans (Hostetler et al., 2013), a new mGluR5 ligand tested in
Animal models of disease
In 7 of the 78 (9%) cases in the sample, the topic was original PET studies of disease models in experimental animals. PET was used to detect increased TSPO expression in a neuroinflammation model in non-human primate (Hannestad et al., 2012) and in a rat study of temporary focal cerebral ischemia (Hughes et al., 2012). Reduced dopamine transporter binding was detected in a rat model of hemiparkinsonism (Fischer et al., 2012), and baseline dopamine D2/3 receptor availability in rat brain
Human diseases
In 7/78 (9%) articles from the sample, the main focus was on PET studies of neurochemical correlates of human disease and substance abuse disorders. Concerning the latter sub-theme, GABA(A) binding was found to be focally increased in brain of smokers (Stokes et al., 2013), and in an application of the competition paradigm noted above, reward-dependent dopamine release correlated with severity of pathological gambling (Joutsa et al., 2012). Phosphodiesterase-IV availability was reduced in
Disease classification and prediction
In 10 of the 78 (13%) articles in the sample, the main topic was disease classification and prediction of progression; with a single exception these were studies of AD. The range of PET methods and structural and functional MR procedures available for the study of AD were extensively reviewed (Reiman and Jagust, 2012). There were reports on longitudinal [18F]-FDG-PET (Gray et al., 2012) and [18F]-FDG-resting state functional connectivity measurements (Toussaint et al., 2012) for stratification
Normal brain function and healthy aging
17 of the 78 (22%) studies from the sample the topic was normal brain function; for 10 of these 17 cases, this was in the context of CBF and [18F]-FDG-PET. CBF measurements in conjunction with TMS and electrophysiology were used to map connectivity of the supplementary motor area (Narayana et al., 2012), and meta-analyses of historical PET and fMRI studies were presented for the study of speech and language encoding (Adank, 2012, Price, 2012), as well as for vestibular processing (zu Eulenburg
Perspectives
It is difficult to separate completely the future of PET molecular imaging in less focussed than is NeuroImage on studies from the future of brain PET as a scientific endeavor. Before drawing general conclusions above the contributions of PET to NeuroImage during 2012, I now present an analysis of the broader historical trends, as derived from a PubMed search of the entire biomedical literature using the search terms PET, fMRI and Brain. Augustine of Hippo said that it is the natural order of
References (81)
Design choices in imaging speech comprehension: an Activation Likelihood Estimation (ALE) meta-analysis
NeuroImage
(2012)- et al.
Oxygen metabolism, oxygen extraction and positron emission tomography: historical perspective and impact on basic and clinical neuroscience
NeuroImage
(2012) - et al.
Cognitive reserve impacts on inter-individual variability in resting-state cerebral metabolism in normal aging
NeuroImage
(2012) The serendipitous discovery of the brain's default network
NeuroImage
(2012)- et al.
Quantitative measurement of cerebral physiology using respiratory-calibrated MRI
NeuroImage
(2012) - et al.
Correlations between FDG PET glucose uptake-MRI gray matter volume scores and apolipoprotein E ε4 gene dose in cognitively normal adults: a cross-validation study using voxel-based multi-modal partial least squares
NeuroImage
(2012) A history of randomized task designs in fMRI
NeuroImage
(2012)- et al.
Positron emission tomography assessment of 8-OH-DPAT-mediated changes in an index of cerebral glucose metabolism in female marmosets
NeuroImage
(2012) - et al.
Naming impairment in Alzheimer's disease is associated with left anterior temporal lobe atrophy
Neuroimage
(2012) - et al.
Can target-to-pons ratio be used as a reliable method for the analysis of [11C]PIB brain scans?
NeuroImage
(2012)
In vivo quantification of dopamine transporters in mice with unilateral 6-OHDA lesions using [11C]methylphenidate and PET
NeuroImage
5-HTTLPR status predictive of neocortical 5-HT4 binding assessed with [1C]SB207145 PET in humans
NeuroImage
The coupling controversy
NeuroImage
Quantitative magnetization transfer provides information complementary to grey matter atrophy in Alzheimer's disease brains
NeuroImage
Alzheimer's Disease Neuroimaging Initiative. (2012) multi-region analysis of longitudinal FDG-PET for the classification of Alzheimer's disease
NeuroImage
Alzheimer's Disease Neuroimaging Initiative. (2013) random forest-based similarity measures for multi-modal classification of Alzheimer's disease
NeuroImage
Identifying improved TSPO PET imaging probes through biomathematics: the impact of multiple TSPO binding sites in vivo
NeuroImage
Combining image-derived and venous input functions enables quantification of serotonin-1A receptors with [carbonyl-11C]WAY-100635 independent of arterial sampling
NeuroImage
Endotoxin-induced systemic inflammation activates microglia: [11C]PBR28 positron emission tomography in nonhuman primates
NeuroImage
Evaluation of [(18)F]MK-0911, a positron emission tomography (PET) tracer for opioid receptor-like 1 (ORL1), in rhesus monkey and human
NeuroImage
Pharmacologic magnetic resonance imaging (phMRI): imaging drug action in the brain
NeuroImage
Mesolimbic dopamine release is linked to symptom severity in pathological gambling
NeuroImage
Non-linear mixed effects modelling of positron emission tomography data for simultaneous estimation of radioligand kinetics and occupancy in healthy volunteers
NeuroImage
Investigating the effects of noise-induced hearing loss on serotonin transporters in rat brain using 4-[18F]-ADAM/small animal PET
Neuroimage
Effects of beta-amyloid accumulation on neural function during encoding across the adult lifespan
NeuroImage
Quantification of metabotropic glutamate subtype 5 receptors in the brain by an equilibrium method using 18F-SP203
NeuroImage
Serotonin-1A receptor binding is positively associated with gray matter volume — a multimodal neuroimaging study combining PET and structural MRI
NeuroImage
Prediction of SSRI treatment response in major depression based on serotonin transporter interplay between median raphe nucleus and projection areas
NeuroImage
Extraction of an input function from dynamic micro-PET images using wavelet packet based sub-band decomposition independent component analysis
NeuroImage
Characterisation of partial volume effect and region-based correction in small animal positron emission tomography (PET) of the rat brain
NeuroImage
Imaging DA release in a rat model of l-DOPA-induced dyskinesias: a longitudinal in vivo PET investigation of the antidyskinetic effect of MDMA
NeuroImage
Dissociable effects of methylphenidate, atomoxetine and placebo on regional cerebral blood flow in healthy volunteers at rest: a multi-class pattern recognition approach
NeuroImage
Mental chronometry
NeuroImage
PET imaging predicts future body weight and cocaine preference
NeuroImage
Quantification of dopamine D(2/3) receptors in rat brain using factor analysis corrected [18F]Fallypride images
NeuroImage
Not quite PIB-positive, not quite PIB-negative: slight PIB elevations in elderly normal control subjects are biologically relevant
NeuroImage
Electrophysiological and functional connectivity of the human supplementary motor area
NeuroImage
Multiple testing corrections, nonparametric methods, and random field theory
NeuroImage
A linear model for estimation of neurotransmitter response profiles from dynamic PET data
NeuroImage
Quantitative PET analyses of regional [11C]PE2I binding to the dopamine transporter—application to juvenile myoclonic epilepsy
NeuroImage
Cited by (11)
Simple and rapid quantification of serotonin transporter binding using [<sup>11</sup>C]DASB bolus plus constant infusion
2017, NeuroImageCitation Excerpt :The correspondingly lower effect sizes in cross-sectional study designs together with the limited sample sizes have resulted in divergent study results. In the case of major depressive disorder, these have transiently obscured effects and raised controversy in the field regarding both, the nature of SERT alterations in depressive patients and the effectiveness of molecular imaging to detect relevant disease markers (Cumming, 2014). In contrast, more recent meta-analytic approaches produced solid effect estimates which can inform future trials on the sample sizes required to detect significant changes in molecular imaging measures (Gryglewski et al., 2014; Howes et al., 2012).
A systematic review of molecular imaging (PET and SPECT) in autism spectrum disorder: Current state and future research opportunities
2015, Neuroscience and Biobehavioral ReviewsCitation Excerpt :The authors found decreased blood flow toward speech-related areas in the left temporal cortex in both adults and children with ASD compared to the control groups (healthy adult males and children with idiopathic intellectual disability) (Boddaert et al., 2003, 2004). While H215O-PET has led to interesting findings of hypoperfusion in ASD, the use of PET to investigate CBF will likely continue to decrease in the coming years (Cumming, 2014) as blood flow can be investigated quantitatively with MR-based methods such as arterial spin labeling (ASL) or dynamic susceptibility contrast, which do not require administration of an exogenous radioactive contrast agent. To date, however, the field of ASD has not yet taken advantage of these techniques, as there is, to our knowledge, only one study that investigated CBF using ASL in children with ASD (Diamandis et al., 2012) and no study that used dynamic susceptibility contrast.
Motor imagery-based brain activity parallels that of motor execution: Evidence from magnetic source imaging of cortical oscillations
2014, Brain ResearchCitation Excerpt :As such, it remains unclear whether ‘pure’ MI elicits the same spatial patterns as ME. Furthermore, the majority of studies examining brain activity underlying MI utilize either fMRI or positron emission tomography (PET) (Hétu et al., 2013), both of which rely on indirect measures of brain activity with low temporal resolution (Sutton et al., 2009; Cumming, 2014). Accordingly, these measures provide rich spatial information, but are limited in their ability to directly measure electrophysiological activity.
Not shooting an elephant
2014, NeuroImage