Molecular imaging of psychiatric disorders has taken a few turns around the lab as a concept, and while some potential imaging methods, namely dopamine transport scanning, have shown some potential, the technique has not really taken off for psychiatric applications due to a mixed bag of results and a limited understanding of the pathophysiology. However, some imaging methods, including an updated version of dopamine imaging, serotonin and norepinephrine imaging, as well as diffusion tensor imaging (DTI), could be applied in select situations such as drug discovery and characterization of disease for not only schizophrenia, but attention deficit hyperactivity disorder (ADHD), chronic depression and other disorders.

A friend once told me he could make out people’s faces in the canopies of trees. This was after he revealed that he was a functional schizophrenic. He also happened to be a brilliant mathematician and programmer. As a medical writer, I wondered what that must look like inside the brain. What are the biomarkers of schizophrenia?

Diagnosis is primarily hinged on observation of a patient’s behavior and his or her documented experiences, but the biochemical seat of schizophrenia, and most psychiatric disorders for that matter, is not very well understood. In fact, researchers have been trying to visualize it via molecular imaging for decades.

Dopamine dreams

Since the early 1980s or perhaps even earlier, scientists have been paying particular attention to dopamine and its transport as a potential target of molecular imaging. You would think that 30-year-old research would have procured a solid imaging technique, but there have been challenges in pinning down schizophrenia and other disorders.

Molecular Imaging Insight discussed the possibility and challenges of molecular imaging for psychiatric disease with Jacob Hooker, MD, PhD, director of radiochemistry for the Athinoula A. Martinos Center, associate director of the PET Core at Massachusetts General Hospital and assistant professor at Harvard Medical School in Boston. He explains that a lot of the neuroimaging studies done in psychiatric disease have had mixed findings.

“Depending on the illness you look at, the given effect may be supported by one finding and then negated by another,” says Hooker. “I think what we are seeing in those cases is a likely consequence of the heterogeneity of mental illness and psychiatric disorders in general. When we call something schizophrenia or bipolar disorder or depression, what are we really ascribing that illness to? We don’t know.”

From the current literature, it may not be possible to image a large group of people and be able to pick out those with a particular psychiatric disease, or one vs. another, much less a screening tool, suggests Hooker.

“This is coming from someone in the field who wants to see that happen, because we need that molecular basis. We need that neuro-structural signature of disease.”

Correlation vs. causation

Many of the scanning techniques tested to date have been largely correlative, not causative. There are several other methods that have cropped up in the research, but this feature focuses on four main mechanisms of molecular imaging: dopaminergic transport imaging, serotonin receptor imaging, norepinephrine receptor imaging, and lastly DTI of white matter in the brain as a back-end visualization of brain function—a means of imaging the consequence of disease. The first three, rather than being causative, are more correlative, and have been dictated mostly by pharmacology research that has shown certain drug therapies to have an impact on disease via these particular proteins or receptors in the brain. However, all of these methods image completely different realms of neurochemistry that happen to be loosely linked to one another. While they are not really comparable, they do all point to some aspect of psychiatric disease even if researchers don’t yet know their role, if any, in disease development.

As a target for psychiatric disease, dopaminergic systems have been as good as any, but it hasn’t been easy for imaging schizophrenia.

“Schizophrenia imaging is a tricky one,” says Shankar Vallabhajosula, PhD, a professor of radiochemistry at the Weill Cornell Medical College of Cornell University in Ithaca, N.Y. “Dopamine receptor imaging has been the most important [imaging method], but its value was never proven.”

Despite the challenges in this particular clinical application, DaT imaging did go on to become a very effective technique for diagnosing Parkinson’s disease, marked by dopamine disorder, and is now FDA approved as DaTscan, also known as I-123 Ioflupane.

Attention seeking scans

In addition to Parkinson’s and other movement disorders, recent research has opened up a small window on ADHD, says Gene-Jack Wang, MD, from the National Institutes of Health and Brookhaven National Laboratory in Upton, N.Y.

One study in particular using F-18 falypride to image nigrostriatal dopaminergic mechanisms underlying attention—specifically D2 and D3 dopamine receptors—showed that treatment with methylphenidate evened out D2 and D3 receptor availability and endogenous dopamine to a point where ADHD subjects were on par with healthy controls. However, the researchers from the University of Cambridge came to believe that dopamine dysregulation is not the ticket to the disease, but rather just one of perhaps multiple factors of ADHD (Brain (2013) 136 (11): 3252-3270. doi: 10.1093/brain/awt263).

Still, the research has been more unclear for other disorders, including schizophrenia. “We have tried to study ADHD and I think it has brought some understanding of the disease, but for schizophrenia it has not been as applicable,” comments Wang.

Where do researchers go from here?

A conversation that began during this year’s Society of Nuclear Medicine and Molecular Imaging (SNMMI) Annual Meeting in St. Louis and continued afterward has provided a more encouraging perspective from Aadrian A. Lammertsma, PhD, vice-chair of research for the department of radiology and nuclear medicine at VU University Medical Center in Amsterdam, Netherlands.

“I believe that we first need to understand what is happening in the brain, i.e. which physiological and/or molecular processes/targets are disturbed,” says Lammertsma. “Here PET could be helpful. Once we have identified processes/markers, the same imaging techniques can be used to stratify patients for the appropriate medication and to assess subsequent response.”

In addition to PET, there is growing interest in using SPECT with I-123 MNI-420. This technique has been explored for neurodegenerative disorders like Parkinson’s and Huntington’s disease, but it could be beneficial for disorders like schizophrenia, bipolar disorder, anxiety disorders and chronic depression. The key to adenosine SPECT is its ability to target A2A receptors in dopaminergic systems in the brain (J Nucl Med 2013; 54:1–8).

Why white matter matters

Current understanding of the symptoms of schizophrenia point to a disconnect between critical regions of the brain. This could be due to developmental miswiring of neural connections and abnormal plasticity in the synapses. Researchers have been looking into this disconnectivity and apparent white matter abnormalities in schizophrenics by adopting DTI imaging.

A recent DTI study of white matter and the effects of clozapine on its integrity in schizophrenics, published in June in Psychiatric Research: Neuroimaging, presented evidence that patients with the disorder had lower fractional anisotropy (FA) in 16 areas of the brain compared to healthy controls. These were the bilateral superior longitudinal fasciculi, inferior fronto-occipital fasciculi, superior and inferior parietal lobules, cingulate bundles, the cerebellum, middle cerebellar peduncles, and left inferior longitudinal fasciculus. Schizophrenics also had higher FA in six areas: the right parahippocampus, left anterior thalamic radiation, and right posterior limb of the internal capsule. This was at baseline. Clozapine was found in previous preclinical studies to have a positive influence on neuroplasticity. Patients in this study who underwent treatment with clozapine appeared to increase FA and alter white matter integrity after 12 weeks of treatment (doi: 10.1016/j.pscychresns.2014.06.001).

But even MRI is unlikely to translate into a screening method for individual patients with mental illness, says Wang. MRI may be less expensive than PET, but it is far more likely that clinics will use blood tests to analyze the neurochemistry of psychiatric patients rather than brain scans.

Approaching mood disorders

Jonathan B. Savitz, PhD, from the Laureate Institute for Brain Research located in Tulsa, Okla., and colleagues published a perspective on psychiatric imaging, and mood disorders in particular, in Molecular Psychiatry last year after contributing to a position paper prepared by the American Psychiatric Association. The conclusion was that no sound means of imaging mood disorders yet exists, but the need to find feasible biomarkers of disease remains very strong. Present therapies are approximations at best, and millions of dollars are wasted on dead-end clinical trials for therapies that don’t seem to be addressing the root of the problem.

“The need for clinical biomarkers has become acute, as their absence particularly has hindered research aimed at developing novel therapeutics,” wrote Savitz et al. “Due at least partly to the lack of well-established pathophysiological targets for new drugs, relatively large numbers of experimental compounds are failing in increasingly expensive late-stage clinical trials. As a result, drug development pipelines are becoming dry, and several companies have discontinued their research and development of pharmaceuticals for psychiatric conditions.”

The authors posited that clinicians have been working mostly in the dark since the development of the Diagnostic and Statistical Manual of the American Psychiatric Association (DSM)-III and many patients never receive a therapy that is adequate to treat their symptoms.

Unlike a scan for amyloid deposition as a sign of Alzheimer’s pathology, which can make a determination of positive and negative predictive value, nothing of the sort exists for mood disorders due to a lack of an objective guidelines for diagnosing psychiatric disease. Savitz and his colleagues focused mainly on functional MRI, but they listed a couple of PET leads that could provide some molecular binding potential. Researchers have noticed a reduction in postsynaptic serotonin 1A in the mesiotemporal cortex of those with major depressive and bipolar disorders as well as an increase in serotonin transporter binding in regions such as the anterior cingulate cortex, the thalamus and insula. However, even these findings have been inconsistent in the literature (Molecular Psychiatry. 2 Apr 2013; 18, 528–539; doi:10.1038).

Nodes of neurochemistry

The techniques mentioned here represent just a slice of the neurochemical targets that researchers have gone after the hopes of imaging psychiatric disease. While looking at neurochemical release and density for PET alone, researchers have looked at serotonin receptor IA and 2A, dopamine D1 and D2 receptors, cannabinoids, and on and on, says Hooker. The problem is that these are all interrelated and it is difficult to know how one affects the other.

“If you modulate dopamine transport, it is going to have some indirect effect on norepinephrine; If you modulate norepinephrine, it is likely to have some sort of effect—I don’t know the pathway, but I’m sure it exists, indirectly, on serotonin 4 or other receptors,” explains Hooker. “Finding out which one is the node and which is one the periphery is hard to know.”

Improving imaging studies

Big funding initiatives have been popping up from country to country to help researchers gain a foothold on the staggering complexity of the brain, but that funding is slated for the basic science of neuroconnectivity and the like, and not on specific disease processes. Even so, says Hooker, this knowledge will most likely lead to the opening of more doors to psychiatric disease and a resurgence of imaging research.

There are some hindering factors in the methodology of molecular imaging for psychiatric disease that could be addressed. Movement and clinical heterogeneity, as well as present and prior exposure to drugs, are major barriers for imaging. Researchers could improve their study models by applying some means of movement and partial volume correction and for accounting for other confounding factors (Molecular Imaging in the Clinical Neurosciences Neuromethods. Volume 71, 2012, pp 305-321).

Keeping the doors open

While there are more questions than there are answers at this point, only additional research can improve what is still an incomplete picture of psychiatric disorder to clarify the potential value of molecular imaging for the countless patients and families affected. 

“In most psychiatric diseases there are no anatomical abnormalities, so I am sure that unraveling the underlying pathophysiology will require molecular imaging,” says Lammertsma. “Both SPECT and PET are feasible, but I prefer PET as I believe that sensitivity and quantification are extremely important.”

As researchers continue to get to know the brain, those most sought after biomarkers that signal cause, and not just correlation, will become more apparent. MII