During the past few years a lot of interest has been generated regarding the possible use of Transcranial Magnetic Stimulation (TMS) as a therapeutic modality for autism spectrum disorder (ASD). The interest stems, in part, from a series of clinical trials conducted in my laboratory in which we used TMS in ASD as a way of correcting a dysfunctional cerebral cortex. Our first trial received the recognition of a EUREKA award from the National Institute of Mental Health. That initial study led us to roughly a decade of research on the subject.

Looking back at our efforts we have had 14 published clinical trials on over 200 patients. Each trial varied in terms of the parameters used for TMS and the particular outcome measures used to establish efficacy.  All of our subjects have been children, usually in the range of 8 to 18 years of age, with high-functioning autism. When dealing with children it is best to take “baby steps” and each change in our protocols has been tweaked so as to insure the safety of our patients. Thus far we have reported quantifiable improvements in several domains of behavior such as irritability and stereotypies. These improvements have been in the range of modest to moderate but still statistically significant overall. In addition, based on teachers’ questionnaires, we have also noted improvements in behavior at school. The latter finding makes a lot of sense as a child that is able to sit still, be attentive, and partake in required educational activities would do better at school than one who is irritable and/or distracted.  Furthermore, the procedure has been well tolerated and side-effects (e.g., headaches) have been minor and transitory. Positive results, when present, are not immediately seen and rely on the plasticity of the brain to correct itself.  Benefits appear to be transitory and, depending on the subject, last approximately 6 months. Booster treatment are necessary to maintain improvement, but each successive treatment appears to be spread further apart.

Based on our experience with TMS we have never seen or reported on a cure for ASD. Non-verbal patients remain non-verbal and any intellectual impairments remain unfazed.  Indeed, our experience with low functioning ASD individuals has been unrewarding. TMS may help correct functional deficits but not anatomical ones.  Cortical malformations and other developmental lesions which may be found in low functioning autistic individuals are not amenable to treatment with TMS.

A recent book by John Elder Robison entitled “Switched On” has idealized TMS treatment as a possible cure for autism. According to the narrative, TMS stimulation allowed John to gain empathy, to subconsciously color his experiences with emotional overtones, and to acquire theory of mind (i.e., he was able to better understand the intentions of others). These effects have been permanent, that is, without further need of booster shots. John participated in several studies at Harvard University none of which was a clinical trial. In the studies, single pulses of TMS were used as a way of investigating specific functions of the brain.  This type of TMS, single pulses, has no discernible effects outside the session of stimulation. Although John’s experiences were both emotional and personal they are best attributed to a placebo effect.

TMS remains an experimental tool in ASD. There haven’t been multicenter trials with large enough number of participants to prove its effectiveness and enable comparisons to other treatments.  Similarly, patients receiving TMS for ASD have not been followed for any considerable length of time. Thus far, the Food and Drug Administration(FDA) of the United States has not recognized the therapeutic value of TMS in ASD.

Properly designed trials will be of importance in order to judge the effectiveness of TMS. A major stumbling block to overcome is the fact that participating subjects can feel the effects of active TMS stimulation. The strong magnetic field not only stimulates the cerebral cortex but also the nerves of muscles in the scalp.  The resultant contraction of scalp muscles cause transitory headaches.  It is therefore difficult to blind participants as to whether they are receiving active stimulation or a sham condition.  Having knowledge of an active treatment, specially one with physical effects like muscle contractions, may promote a strong placebo response, just as in the case of John Robison.

Needless to say that much more research is needed in regards to TMS and ASD. In our laboratory we are presently trying to answer a few important questions.  First, why do ASD patients respond in markedly different ways to TMS, some improving while others remain impervious to the intervention?  In general, individuals with ASD have an imbalance in their autonomic system, i.e., the part of the nervous system that controls involuntary functions of the body such as heart rate. TMS is usually applied to areas of the frontal lobes that help regulate the autonomic system.  It seems likely that the improvements observed with TMS regarding irritability, anxiety and attention may be the result of modulating these key areas of the brain.  We are now interested in using measurements of autonomic reactivity (e.g., heart rate variability, galvanic skin response) as possible predictors of outcome for ASD.  Second, we have observed a synergistic effect when combining TMS with neurofeedback. It may be that a better therapeutic effect can be obtained when combining TMS, with other interventions, e.g., neurofeedback, drug therapy. Lastly, several of our clinical trials suggest that participating patients showed improvements in attentional control. Not surprisingly, this executive function is controlled by the frontal lobes. It may be that the improvements noted in our clinical trials may have been due to a subset of patients that had attention deficits.  If so, TMS therapy would best be served by targeting this subgroup of patients.

In summary, clinical trials using TMS in ASD have been few with modest to moderate results. Positive outcomes have been noted primarily in high functioning individuals. Despite the popular press, no reports of “cures” have been reported.  Well-designed large-scale clinical trials are still needed to judge the effectiveness of this intervention.  The few cases of seizures reported while using TMS should draw attention to the fact that we need to be cautious when treating ASD patients, a condition where one third of individuals suffer from epilepsy. In regards to TMS, be cautious of commercial enterprises charging a lot of money and offering cures or improvements of any or all autistic symptoms. Patients and their parents should be aware of the positives and negatives of TMS therapy before investing a significant portion of their financial resources to this intervention.

Addendum 3/1/17: In a previous year I had met with Lindsay Oberman who had clarified to me that none of the studies in which John Robison participated were clinical trials. My understanding was that John had received single stimulation in mechanistic studies designed to study brain plasticity. John clarified that none of these single-pulse studies had any effect on him. The study which had long lasting cognitive effect on John used 1 pulse per second for 30 minutes over Broca’s area. I appreciate John’s clarification. (I must clarify however, that I still believe that John’s effect were placebo induced- see comments below. Also the practice of re-using the same patients in multiple research trials is highly questionable).