UCLA's Dr. Allan Wu Discusses Transcranial Magnetic Stimulation (TMS)


The Parkinson Alliance and Team Parkinson have been funding some leading edge projects at UCLA for the past several years.  Dr. Allan Wu, MD is doing fascinating work in the area of TMS—Transcranial Magnetic Stimulation.  Recently, Dr. Jeffrey Wertheimer, Clinical Neuropsychologist from The Parkinson Alliance staff, interviewed Dr. Wu about his work.  Below is the full interview.

Dr. Jeffrey Wertheimer:  Could you give a brief overview about Transcranial Magnetic Stimulation therapy? What are the hypotheses as to how it can benefit individuals with PD?

Dr. Allan Wu:  Transcranial magnetic stimulation (TMS) is a non-invasive method of stimulating the brain.  A coil of wire, insulated in plastic, is put on the head like a big spoon.  Each TMS pulse then produces a magnetic field that stimulates the surface of the brain beneath the coil.  Since magnetic fields do not hurt, this process is not painful.

When we use TMS for therapy or treatment, we apply a series of pulses over each part of the brain we want to stimulate.  A series of pulses given at a time is called repetitive transcranial magnetic stimulation (or rTMS).  A typical treatment with rTMS consists of daily sessions (5 times a week) for 2 to 4 weeks.  Each session lasts about an hour with stimulation lasting between 30-45 minutes.

This is a growing and exciting concept, particularly since rTMS is now an FDA approved therapy for certain types of major depression.  In depression, rTMS is applied for 4-6 weeks over a part of the brain within an area called the prefrontal cortex.  After 2 weeks, patients who receive rTMS show an significant improvement in depression compared to patients who receive a placebo (sham-rTMS). 

We are currently studying rTMS to see if might be a clinical treatment for Parkinson’s disease (PD) symptoms in a study that is funded by the Michael J Fox Foundation (the MASTER-PD trial).   This trial builds off the excitement and possibility that rTMS is now an FDA-approved treatment for depression.

In our MASTER-PD trial, we study the effects of 2 weeks of rTMS on motor and non-motor symptoms of Parkinson’s disease.  The motor symptoms we aim to improve include walking, freezing of gait, imbalance, rigidity, loss of dexterity, and slowness of movement.  The non-motor symptoms we aim to improve are those related to depression (lack of motivation).  Patients will be followed up to 6 months after rTMS to see how long effects of rTMS may last.

The first hypothesis (or question) we want to test is whether rTMS over the prefrontal cortex (the part of the brain that is stimulated when rTMS is used to treat depression) would improve depression symptoms in PD patients.  Because rTMS over the prefrontal cortex is designed for depression symptoms, this type of rTMS might not be expected to help the motor (movement) symptoms in PD.  However, just like improvement of depression with antidepressants sometimes helps movement symptoms, the possibility still exists that rTMS targeted toward depression may help movement symptoms too.
The second hypothesis is whether rTMS over the motor cortex (the part of the brain that most directly controls voluntary movement) would improve movement symptoms in PD patients.  Because rTMS here is directed toward movement, this type of rTMS may not help depression symptoms.

To answer these questions, we designed our clinical trial to randomize each participant to get either rTMS over the prefrontal cortex (for depression symptoms) and/or the motor cortex (for movement symptoms). 

The way we do this is that the participant will get something that feels like rTMS over the prefrontal cortex.  For half of the participants, this will be real rTMS and for the other half of participants, this will be sham (placebo) rTMS which feels like rTMS but does not actually stimulate the brain. 

Every participant will also get something that feels like rTMS over the motor cortex.  For half of the participants, this will be real rTMS and for the other half of particpants, this will be sham (placebo) rTMS.

In the end, on average, out of every 4 participants, one will get rTMS over prefrontal cortex; one will get rTMS over motor cortex, one will get rTMS over BOTH prefrontal cortex and motor cortex; and one will get sham-rTMS over BOTH prefrontal cortex and motor cortex.  This means every participant will have a 75% chance of receiving a form of real rTMS over one or both brain regions and a 25% chance of receiving a sham rTMS. 

It is important to realize that the participants in rTMS clinical trials who get purely sham-rTMS condition always show a large benefit from receiving sham-rTMS.  It is just that real-rTMS seems to improve patients even more than sham-rTMS.  Fortunately, side-effects in all rTMS clinical trials in Parkinson’s disease and depression patients have been minimal and seems to be well-tolerated.

The reason why rTMS works is not entirely clear.  This is one aspect about rTMS that our laboratory at UCLA is particularly interested in.

One simple way to think about what we think we are doing with rTMS is to imagine that rTMS stimulation of your brain for 30 minutes a day (a session of TMS) is like challenging a brain circuit to adapt to a new circumstance.  If you were to practice a new skill (like a tennis swing) for 30 minutes for 1 day (1 session), it may not change much overall in your skill or your brain.  However if you continue to practice this skill for 5 days every week for 2-4 weeks, you probably would acquire a new skill.  This means your brain has changed and adapted to incorporate this new skill which hopefully will become more second-nature to you.  In a similar way, we are hopeful that challenging certain brain circuits with brain stimulation for 5 days a week for 2 weeks will have the effect of causing the brain to re-adapt to the PD condition in a more beneficial way which will make either movement or mood easier for the patient to control.

Dr. Wertheimer:  Can you describe the inception of your research with TMS and PD (Stimulation Trials In Movement Disorders or STIM-PD)?

Dr. Wu:  I first started using transcranial magnetic stimulation (TMS) while an Assistant Professor in the Movement Disorders program at USC around 2003.  Our earliest projects used TMS as a tool to measure the excitability of the human motor system.
What this means is that we would stimulate the movement part of the brain (the motor cortex) with one pulse of TMS at a time and cause a muscle in the hand to twitch.  We would measure the muscle twitch with an electromyogram (using electrodes taped to the skin).  By changing the intensity of the TMS pulse, we could change the size of the muscle twitch.  This is a way to measure how “connected” the motor cortex (brain) is to the muscle that we are measuring.  This measurement is what we call “excitability.”  We can use this measurement of excitability to find differences between PD patients and subjects who did not have PD.  This proved to be a useful tool to measure how having PD changes the “excitability” or “connectivity” between the brain and hand muscles.  This tool could then be used to see how well patients responded to various treatments for PD.  However, this tool mainly tells us something about PD (like a brain scan or x-ray), but does not take advantage of the fact that TMS might be able to directly stimulate the brain in a way that might help their symptoms.

When I moved to UCLA in 2005, I established the UCLA Motor Control Laboratory and started to design repetitive transcranial magnetic stimulation (rTMS) studies.  Repetitive TMS is when a series of TMS pulses are given repetitively over one part of the brain at a time for 15-30 minutes at a time.  This can be done at low- (less than 1 pulse per second) or high-frequency (5-10 pulses per second).  When you deliver one set of rTMS in this way, it leaves a measurable after-effect that lasts up to 15-30 minutes.  This means that rTMS causes the brain to adapt to the stimulation and we think that this adaptation may occur in a way that could be benefiical for symptoms of PD.
We started by testing different kinds of movement tasks.  Our lab demonstrated some of this early work years ago to members of The Parkinson Alliance and Team Parkinson in Los Angeles.  These tasks involved repetitive tapping of keys and moving of levers between targets of varying size, direction, and distance.  Our goal was to explore how the after-effects of rTMS stimulation could affect the control of movement in patients with PD and control subjects without PD.  We wanted to see if different ways of delivering rTMS might help improve some of movement control in PD patients.
These studies became the STIM-PD (Stimulation Trials In Movement Disorders-PD) protocols.  We started to collect data from PD patients who would receive different forms of rTMS and have their ability to perform movement tasks tested before and after rTMS.  We also tested patients both ON and OFF their medications.  These were single-visit studies where subjects would visit the lab, receive one type of rTMS and be tested on one or two movement tasks. These were exploratory studies with results that led to two NIH grant submissions which were not funded.  In retrospect, although we had some promising results, the exploratory nature of our studies, the different types of rTMS, and short-duration of each rTMS (single-visit) did not lead to definite conclusions.

Around this time, we were using the STIM-PD infrastructure to investigate the idea that we could use rTMS over different parts of the brain to address different symptoms.  This was a concept that a group of us who study rTMS and PD published in 2008.  We then designed the STIM-PD protocol that I discussed at a 2008 Team Parkinson pre-LA Marathon talk where we would see if rTMS over the motor (movement) cortex could change how a patient was able to perform movement tasks; and also see if rTMS over a part of the brain controlling mood (prefrontal cortex) could change how the same patient performed on mood tasks. 

By this time, rTMS had been FDA-approved for the treatment of depression.  Unlike the STIM-PD protocols, the way rTMS is used for depression is to apply rTMS (prefrontal cortex) on a daily basis for 2-6 weeks.  This was an exciting development because it proved that rTMS could have a role in the treatment of neuropsychiatric symptoms (like depression) but had to be applied in a sustained (day-after-day) way to have its effect.  This opened the door to apply this method of daily rTMS to PD patients as well.

A group of us then combined these concepts and designed a proposal to use rTMS to treat PD mood (depression) and motor (movement) symptoms to NIH and the Michael J Fox Foundation (MJFF).   In this proposal, we proposed a multicenter clinical trial in PD patients that would test whether rTMS over the prefrontal cortex (just like the FDA-approved rTMS for depression) would treat depression symptoms in PD.  At the same time, we included conditions for subjects to get rTMS over the motor cortex to treat movement (motor) symptoms in PD.   This study design, comparing prefrontal vs motor cortex stimulation, in part, resulted from the UCLA experience in the STIM-PD prtoocols.

When we designed this clincal trial, the administrative site was assigned to Beth Israel Deaconess Medical Center at Harvard given their extensive experience with the use of rTMS in many neurologic diseases.  The statistical site was assisgned to the University of Florida, Gainesville given their recent work on rTMS on anxiety in PD patients.  At the time, the STIM-PD protocols were the extent of rTMS experience at UCLA.  Thus, the STIM-PD work served as a practical demonstration that UCLA should be included in this clinical trial, and helped justify the study design (mood vs motor symptoms) that was proposed. This clinical trial project was eventually funded by MJFF as the MASTER-PD project.

Dr. Wertheimer: 
You had mentioned in a note to Carol Walton, CEO of The Parkinson Alliance, that the “seed money” initially provided to you for your research endeavor was instrumental in your research. Could you elaborate on how Team Parkinson and The Parkinson Alliance contributed to the growth of the clinical TMS research in PD?

Dr. Wu:  We received early support shortly after arriving at UCLA when we were just launching our rTMS protocols (the STIM-PD work).  The existing equipment at that time was only capable of generating brief trains of 5 Hz before they would overheat.  Pilot funds were used to purchase power boosters and another air-cooled coil that would allow the sustained 5 Hz rTMS trains which were used in the STIM-PD trials.  Funds were also used to repair and service one of our early TMS units and to pay subjects for their time.  Eventually, these projects were the stepping stone to our collaboration with other academic centers in our current rTMS clinical trial which was then used to justify the purchase a new generation of TMS equipment that now is capable of faster (10 Hz) stimulation and a sham-coil that is being used in the current MASTER-PD clinical trial.

More recent seed support has allowed us to support Janice Lin’s time to invest in the development and application of a brain mapping protocol using MRI scans to our clinical trial.  We are now using Dr Lin’s MRI scanning methods to see if we can find out how rTMS might work to help PD patients. These scans allow us to see which parts of the brain are active when PD patients are trying to do certain tasks (like press buttons) or just laying still.  By comparing the patterns of brain activity before and after the 2 weeks of rTMS in our clinical trial, we can see what kind of changes a daily application of rTMS produces in brain function.  We have now scanned 4 of the 5 patients we have enrolled in our trial at UCLA.  More details about this protocol will follow in a later question below.

Dr. Wertheimer:
  What other grant support have you received?

Dr. Wu:  We have had 3 major funding sources.

The first is a 5 year Howard Hughes Medical Institutes Early Career Award which runs through 2012.  This award funds my laboratory staff, laboratory infrastructure, early work with motor tasks, and current exploratory brain stimulation projects in both Parkinson’s disease and dystonia.

The second is the Michael J Fox Foundation. In 2009, they funded our multicenter clinical trial of rTMS for treatment of mood and motor symptoms in PD.  Information about this grant can be obtained at:

Our third grant was from the CurePSP Foundation.  They funded a 2 year project using the STIM-PD infrastructure, but as applied to patients with atypical parkinsonism – particularly progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD).

Dr. Wertheimer:
  What centers are involved in your clinical trial?

Dr.  Wu:
  Currently 5 centers:
1. Beth Israel Deaconess Medical Center, Boston, Massachusetts.  (PI: Alvaro Pascual-Leone, M.D., Ph.D.)
2. UCLA (PI: Allan Wu, M.D.)
3. The Cleveland Clinic (PI: Hubert Fernandez, M.D.)
4. University Health Network, Toronto (PI: Robert Chen, BChir, MA, MB, MSc)
5. University of Florida, Gainesville (PI: Aparna Wagle-Shukla, MD)

Contact info is at:  http://clinicaltrials.gov/ct2/show/NCT01080794

Dr. Wertheimer:  What are the inclusion and exclusion criteria for your multi-center study?

Dr. Wu:  A lot of detail about this study can be found at our clinicaltrials.gov listing:

Summary inclusion criteria:
1. Diagnosis of Parkinson’s disease (PD) approximately 3 years – on treatment with either levodopa and/or a dopamine agonist
2. Movement symptoms in spite of stable PD medication treatment.
3. Depression symptoms (including current or past treatment with an antidepressant).  Potential subjects will be interviewed by our clinical psychologist or psychiatrist for purposes of this criteria.

Summary exclusion criteria:
1. Metal in the head (dental fillings are okay)
2. Implanted pacemaker, medication pump, deep brain or other nerve stimulator, or ventriculoperitoneal shunt
3. History of seizures or unexplained loss of consciousness
4. Family history of medication-refractory epilepsy
5. Significant suicidal ideation (or hospitalization for suicide attempts)
6. History of ECT (electroconvulsive therapy)
7. History of bipolar disorder or psychosis (not related to medications)
8. Unstable medical conditions
9. Dementia
10. Intolerance of being OFF medication for 12 hours for testing at baseline and follow-up visits
11. Substance abuse or alcoholism

Dr. Wertheimer: 
How many participants have enrolled?

Dr. Wu: 
UCLA has enrolled 5 patients (and excluded approximately 3 times this number).  The total enrollment to date across all sites is 20 patients.  We are shooting for a goal of 120 patients by the end of the study.  We are in the 2nd year of the 3 year project and anticipate an extension of the study into a 4th year to complete our data collection.


Dr. Wertheimer:  What is the current status of your research endeavor?

Dr. Wu:  Overall, MASTER-PD is enrolling at below a 50% pace.  So, this is a priority project for us.  We are increasing our enrollment with the benefit of MJFF Foundation initiatives for increasing recruitment, sending mailers to local physicians not limited to PD patients, have plans to outreach more into psychiatry clinics, and have posted a recent LA Times advertisement.  We are also adding a 6th site in the northwest United States to increase our enrollment.  All site PI’s have spoken regularly and remain very positive about the outcome of the trial.

We are also encouraged that Beth Israel Deaconess Medical Center (Boston) has now expressed interest in our add-on MRI scanning methods to find out how rTMS works.  We are hopefully going to set up a means to collaborate on our MRI scanning protocol within the MASTER-PD clinical trial.

We recognize that we are excluding a number of patients with PD who would be candidates for rTMS for either depression or motor symptoms, but just don’t have both symptoms. 

For the motor symptoms, Dr Choi Deblieck (our TMS Fellow and Coordinator for the MASTER-PD trial) has worked with me to develop a new protocol in association with our neurorehabilitation colleagues that may lead to the ability to include patients with motor (movement) disorders (including stroke and PD) to receive multiple sessions of rTMS.  This protocol was just approved by our UCLA IRB (ethics board) and has not yet started.

For depression symptoms, we have been in touch with other rTMS research groups at UCLA which study depression and anxiety, but unfortunately, those protocols currently exclude PD patients right now.

On a broader sense, our lab remains interested in neuromodulation for patients affected by movement disorders.  and we have other exploratory projects underway looking at other ways of using rTMS or other technologies for the non-invasive stimulation of the nervous system for possible clinical benefit.

Dr. Wertheimer:
  Has TMS therapy been effective in treating symptoms of PD? If yes, which symptoms are most benefitted? Which symptoms do not tend to benefit from TMS?

Dr. Wu:  From the STIM-PD trials, we did not show consistent results on movement tasks, but we used so many different ways of measuring change in movement control, that we could not draw any definite conclusions.  However, I can say that there was never any deterioration or worsening and that there was frequent comments of improvements for several days after stimulation.  This could reflects a placebo effect and the positive outlook of patients who participate in exploratory studies just as much a definite magnetic stimulation effect.  As such, we really do need a sham-controlled study to find out if it is the magnetic stimulation that can produce these effects.

We cannot comment specifically on the response of any our PD patients who have participated in the MASTER-PD study thus far.  I feel there are clearly clinical effects and thus far we have not seen any signfiicant adverse effects. 

It is also important that the MASTER-PD clinical trial is based on some solid data for the benefits of rTMS.  The data is strongest for the depression symptom as we are using rTMS stimulation settings that are based on those that were used to show the FDA that rTMS for 4-6 weeks significantly improved depression symptoms for 3-6 months compared to placebo (sham) rTMS stimulation.  Smaller studies have suggested that rTMS to prefrontal cortex can help depression in PD patients.  This study is the largest sham-controlled study in depressed PD patients to date.

There are many small studies of rTMS that suggest that motor cortex stimulation in PD will help motor symptoms in PD.  However, there has not yet been a large multi-center study to confirm this finding.  Our multi-center study will thus contribute to this finding out if these smaller studies were on the right track or not.

Dr. Wertheimer: 
Does TMS therapy impact non-motor symptoms related to PD?

See above answer with regards to depression.

Dr. Wertheimer: 
Since fMRI scans have been added to your study, have you been able to identify the mechanisms by which the TMS is effective? Moreover, have you been able to determine how TMS benefits individuals with PD? If so, can you elaborate?

Dr. Wu:  It is clearly too early to say anything definitive about the results from this project, particularly since we have not unblinded the data yet for analysis purposes.  We have scanned 4 of our first 5 subjects enrolled in the MASTER-PD study.  These 4 subjects all tolerated the scan, but one felt that the protocol was a bit long.

In this project, we are taking each patient enrolled in the MASTER-PD project and asking each patient if they would be willing to undergo a special brain MRI scan, called a functional MRI scan, at the beginning (before) and at the end (after) the 2 weeks of rTMS.  This kind of functional MRI is just like a regular MRI scan of the brain in that we get a picture of each person’s brain.  However, we also ask each patient to perform some movement tasks while pictures are being taken of the brain.  These tasks are button pressing tasks in response to cues presented on a video screen in the scanner.  With a functional MRI, we can see which parts of the brain are active in order for patients to do these tasks.  We have one task that emphasizes movement itself (motor system) and another task that emphasizes control of movement (prefrontal system).  We can take each patient’s brain and compare the pattern of activation near the beginning of getting rTMS to the end of the rTMS to see if the rTMS has caused any changes in the brain pattern.

Because this is being done in a clincial trial which randomizes each patient to get either real or sham stimualtion over the motor cortex or real or sham rTMS over the prefrontal cortex, after we have a few more patients in each group (we don’t know yet which group our existing patients are in right now), we can start to see if there is any hint toward knowing how rTMS over several days over either motor cortex or prefrontal cortex affects how the brain works on these tasks.

We have learned that it is not practical to scan all patients ahead of time and after the rTMS because of logistical concerns of making 2 more trips (for MRI scans) in a study that is already quite demanding on patients’ time.  So we have started to scan during one of the first 3 days of rTMS and during one of the last 3 days of rTMS.  Each scan is conducted PRIOR to the rTMS session because otherwise the scan would pick up immediate effects of rTMS and not the overall effects of multiple days of rTMS.

We have been pursuing this project with the knowledge of the other sites.  The protocol has attracted attention from Beth Israel Deaconess Medical Center (BIDMC), the coordinating site for MASTER-PD.  As such, we will be sharing our scanning protocol with BIDMC.  This is another good example of seed funding starting at UCLA that now may enhance the outcomes from this trial.

Dr. Wertheimer:
  Have you had any unexpected findings in the analyses that you have conducted thus far?

Dr. Wu:  No.  No analysis has been done to date. 

Dr. Wertheimer:  What is your vision for this research? What are the short and long-term goals at this point in time?

Dr. Wu:  The overall focus of attention in the lab now is the MASTER-PD study.  We are remain very excited about the study even though recruitment has been slower than anticipated.  We see 3 very important short-term outcomes from this study.  First, we may discover how much and whether rTMS may be helpful for PD symptoms in patients with PD.  This may open the door to larger clinical trials that are more inclusive of PD patients and can enroll in a more broad forum such as the translational clinical trials networks that NIH is currently building.  Second, we uniquely will be able to tell if there is specificity in where rTMS is placed.  There has not been a previous multicenter study that compared two different ways of applying rTMS within the same study in a placebo/sham-controlled way.  If we are able to confirm our initial thoughts, we would be able to tell a future PD patient that if the problem with medication is that if they cannot adequately control depression symptoms, then we should apply the prefrontal rTMS, but if the problem with medications is that they cannot adequately control their movements, then a motor cortex rTMS may be more appropriate.  Third, the MRI data we are collecting will be important basic science data that will be useful for discovering how rTMS exerts its effects on the brain which can be interpreted within the context of the study.

On a more broader goal, my lab is interested in methods of brain neuromodulation and we are working on some exploratory projects in collaboration with other groups (not always with PD) on alternate methods of brain stimulation such as direct-current stimulation and trigeminal nerve stimulation.
I believe there will always be a niche for methods for patients to help manage symptoms that fall between medication therapy and an invasive option such as DBS.  It is one my goals to figure out what is not functioning efficiently in a brain network and help it run more efficiently.  The fact that levodopa and DBS (as examples of effective treatments) are able to correct the function of the brain in PD (and help patients) is a hint that there are likely many avenues to help “modulate” PD symptoms.  We are still looking for the reasons and rules as to how the brain in PD processes information differently and controls movement differently than people without PD.  I think when we look back, we will find that rTMS, at best, will become known as one of these alternative therapies for some patients, but that there will be smaller and more efficient alternatives in the future.  It is my hope to be at the forefront of these neuromodulatory therapies, particularly the non-invasive ones, and also to be there trying to figure out how it is that these methods do their work for the benefit of patients.

Click for a printer friendly version