In TMS research and clinical treatment, the Dorsolateral Prefrontal Cortex (DLPFC) is the equivalent of a “laboratory mouse” as it is everyone’s target region to stimulate. But why? Is there a good reason to use TMS over DLPFC to improve symptoms of Depression, Anxiety, Addiction, Headaches, Cognitive Impairments, and Autism just to mention a few? Critics use this line of thinking to suggest that what we are doing is equivalent to snake oil, advising that no one treatment could possibly work for such different conditions. We don’t use the same medications to treat these disorders, so why would we expect that stimulation of one brain region would work? Is it simply that we know we can modulate this region, so all things being equal, at least we know it is safe and feasible? Is it that we truly believe that dysfunction of the DLPFC underlies all of these conditions? Is there something special about the DLPFC insofar as it represents a “doorway” into the pathophysiological networks associated with all of these conditions? Or does it not matter what region we stimulate because the mechanism of action of TMS is not as focal as we think and we can stimulate anywhere and see effects? I would argue that it is not “snake oil” nor a “cure all”, but there is good reason to stimulate DLPFC for a range of disorders..
Early brain imaging studies in depressed patients found that there was a decrease in activity in the left DLPFC and in its connectivity to limbic structures including the amygdala, anterior cingulate, and insula. These findings have held up and have been further supported by data suggesting that successful treatments for depression including brain stimulation, psychotherapy, and pharmacological treatment are associated with an increase in activity and altered connectivity between DLPFC and limbic structures. Early basic physiology studies also suggested that high-frequency TMS resulted in increased cortical excitability. This has become less consistent where recent literature highlights both inter-individual and intra-individual (state-dependent) variability in response to TMS.
In the late 1990’s pioneers in TMS had good reason to try high-frequency rTMS to left DLPFC as a treatment for depression. And it worked! One thing I’ve learned in science is when you are trying to come up with a new idea, you look at the existing literature, you then design your study based on that literature, you make predictions (or hypotheses) and sometimes things work out the way you think and sometimes they don’t. If they work out, it may or may not be through the mechanism that you thought, but if your studies do work the way you predicted, you follow it up with more studies and others try to replicate your findings. When there is something replicable and reliable in the literature, the excitement and generalized application of the finding tends to snowball. This is not unique to TMS but happens across many scientific fields. But in TMS after many studies showed that high-frequency TMS to left DLPFC improved depression and the FDA cleared the first TMS device for commercial use, there were soon hundreds of thousands of people lining up to receive TMS, and not just for depression.
Independent of the hype, it is clear that high-frequency TMS to left DLPFC does seem to improve depressive symptoms. But, just because it works, doesn’t mean that other regions won’t work also. In fact, both dorsomedial and orbitofrontal (ventromedial) prefrontal cortex are also strongly connected with limbic and subcortical reward circuits and are effective TMS treatment targets for some individuals with depression. However, since currently there is not a way to predict who may respond best to which targeted treatment, and the other targets are not FDA cleared nor covered by insurance, most clinics do not yet offer alternate protocols.
If we think TMS to DLPFC works for depression and it is safe and tolerable, why not try it on other disorders? This thought has crossed the minds of many doctors and researchers over the years. Certainly, not all disorders are related to dysfunction of the DLPFC, but the DLPFC may be the most highly connected region of the cortex. It is connected not simply to the limbic system, but also to other prefrontal and parietal regions as well as subcortical deeper structures. It plays a critical role in cognitive and reward processing. Additionally, many disorders share pathophysiology as well as clinical symptomatology. In other words, Neurological and Psychiatric Disorders are not as distinct as the DSM would have you believe. One should also note that DLPFC encompasses a pretty large chunk of brain with several nuclei that have specific connectivity patterns. For example, DLPFC is critical for executive functioning skills including spatial attention, task switching, and working memory. In this way, it is not surprising that many studies across multiple disorders including Schizophrenia, Dementia, and Depression have found high-frequency TMS improves working memory and other Executive Functioning skills. As a side note, TMS does not appear to be able to improve these skills in otherwise healthy individuals. This may be due to ceiling effects whereby there is a limit to how efficiently the information can be processed in the brain and most of us work at peak efficiency already. So TMS will not make you “smarter”. In the case of depression and anxiety, similar to cognitive behavioral therapy (CBT) which helps patients modify how they think about their emotions and teaches them behavioral strategies, TMS to DLPFC may improve depression, anxiety, irritability, and other chronic pain conditions via similar (“top-down”) mechanisms. The DLPFC is a hub within the reward circuitry and imposes top-down control of subcortical dopaminergic structures implicated in addiction. Thus, modulating excitability and connectivity between these structures may also treat addiction.
With all this being said, TMS to DLPFC does not work for everyone and does not work for every condition. TMS is a brain stimulation tool. It is not a “happy pill” or a cure-all. As I’ve discussed in previous blogs, TMS’s effect depends on stimulating the “right spot” for each person in the “right way”. So, what about this idea that the mechanism of action of TMS is not as focal as we think and we could stimulate anywhere and see effects? This may also be true for some people. Often when randomized trials employ “placebo” or “sham” conditions, the assumption is that you haven’t done anything to the person’s brain, so the person couldn’t have actually gotten better. This is far from the truth. If someone gets placebo or sham TMS they are still coming into the center every day. That means they have to get up, get dressed drive to the center and spend an hour or more with a kind, sympathetic TMS operator who will ask them how they are doing and engage with them socially. This in and of itself is therapeutic. Then, they will receive 30 or more minutes of “treatment”. This treatment will not include a magnetic field, but it still “stimulates” the brain via sound and tactile pulses that are processed by the brain. What is the effect of 10 Hz clicks or taps on your head? I can’t say for sure, but there is reason to believe that there is some general effect of sensory stimulation. In every randomized trial, we see some proportion of participants who have a clinical response in the sham group. This placebo response is not a “made up” effect, but rather one that is real to the patient and the clinician. We may not understand why they are better, but I would argue we don’t understand all that well how drugs work, or how active TMS works, or for that matter how neural pulses lead to thoughts and behaviors and emotions even in the healthy brain.
So, what does this mean? Do we continue to stimulate DLPFC? Sure. It is safe, it is effective (for many individuals with specific conditions), we know it is close enough to the surface of the scalp to respond to the magnetic field and we know the area we are treating is connected to relevant regions and networks. But this does not mean that we should stop looking for other targets which may be more effective or expect it to work for all conditions? While I believe that there is a general effect of coming into the center and having a device tapping and clicking on your head, I also believe that targeting that magnetic field to the right place in the right way is also critical to optimizing the effect. As with all of medicine, I have no doubt that 20 years from now the field will have developed to a point where we will look back and recognize how much we didn’t know and how relatively unsophisticated our tools and techniques were (are) at this time. Until then, we must continue to push the envelope and develop evidence-based TMS treatments for a range of neurological and psychiatric disorders.