Anxiety, Anxiety and Fear, Fear
Conditions
Brief summary
Given the overall lack of treatment adherence/efficacy, side effects of drugs, and the substantial burden of anxiety disorders on the individual and on the national healthcare system, there is a critical need for mechanistic research into the CNS mechanisms that underlie these disorders. Accordingly, the objective of this grant is to use noninvasive neuromodulation to causally identify the key neural mechanisms that mediate the cognitive symptoms of anxiety. This project is relevant to public health because it has the potential to lead to novel repetitive transcranial magnetic stimulation treatments for pathological anxiety.
Detailed description
Although extensive research has explored the involvement of subcortical structures in arousal, arousal symptoms are only one facet of the symptom profile shared across anxiety disorders. Much less is known about the cognitive symptoms (i.e. difficulty concentrating) experienced by anxiety patients. Accordingly, there is a critical need for mechanistic research into the CNS mechanisms that mediate the cognitive symptoms experienced by anxiety patients. Without such research, treatment development for these disorders will continue to make slow progress. The objective of this application is to determine the key neural mechanisms that mediate the cognitive symptoms of anxiety. The central hypothesis is that the right dorsolateral prefrontal cortex (dlPFC) regulates emotion through top-down inhibition of emotion-related regions. The approach will be to use repetitive transcranial magnetic stimulation (rTMS) to study the effect of right dlPFC activity on objective and subjective measures of induced anxiety, anxiety-related working memory deficits (WM), and TMS-evoked blood oxygenation-level dependent (BOLD) responses during simultaneous TMS/fMRI (i.e. target engagement). The rationale for this approach is that by experimentally manipulating right dlPFC activity using rTMS, this research will be able to causally demonstrate involvement of this region in anxiety regulation, which could translate to future targeted rTMS treatments for anxiety. The first aim will be to determine the effect of a 1-week course of rTMS treatment (1 Hz vs. 10 Hz; right dlPFC target) on anxiety using the threat of unpredictable shock paradigm. The second aim will be to determine the effect of a 1-week course of rTMS treatment (1 Hz vs. 10 Hz; right dlPFC target) on anxiety-related WM-deficits using the Sternberg WM paradigm during threat of shock. The third aim will be to demonstrate target engagement by measuring BOLD responses evoked by TMS pulses to the right dlPFC during threat of shock. The work is innovative because it will combine advanced neuromodulatory techniques (fMRI guidance, electric-field modelling, neuronavigation, active-sham control) with a translational threat of shock paradigm. PUBLIC HEALTH RELEVANCE: Once completed, this research should yield direct evidence for a causal role of the right dlPFC in anxiety regulation, complete with evidence of target engagement, and a novel application to anxiety.
Interventions
DEVICErTMS to the right dlPFC
A Magventure MagPro 100X stimulator with a B65 active/placebo figure-8 coil will be used. The TMS coil will be placed on the head over the target. rTMS intensity will be 100% of the motor threshold (MT), adjusted for field strength difference at motor cortex and target cortex using the individual E-field model. Subjects will receive 3000 pulses/session.
Sponsors
University of Pennsylvania
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
BASIC_SCIENCE
Masking
TRIPLE (Subject, Investigator, Outcomes Assessor)
Masking description
Both the subject and the TMS operator will be blinded as to the study condition (active vs. sham)
Intervention model description
Aims 1 and 2 will be tested using a between-subjects design where 1 group of healthy volunteers will receive 4-day courses of active and sham 1 Hz stimulation to the right dlPFC, while the other group will receive 4-day courses of active and sham 10 Hz stimulation to the right dlPFC. Aim 1, will test the effects of this stimulation on anxiety during the NPU threat task. Aim 2, will test the effects of this stimulation on anxiety-related working memory (WM) deficits using the Sternberg WM paradigm. Aim 3 will be tested using a within-subjects design where this same group of subjects will receive single-pulse active and sham stimulation to the right dlPFC during the Neutral, Predictable, and Unpredictable (NPU) threat task while in the MRI scanner. BOLD activity to the TMS pulses will be the primary outcome measure.
Eligibility
Sex/Gender
ALL
Age
18 Years to 50 Years
Healthy volunteers
Yes
Inclusion criteria
* Subjects must be 18-50 years old * Able to give their consent * Right-handed
Exclusion criteria
* Non-english speaking * Any significant medical or neurological problems * Current or past Axis I psychiatric disorder(s), active or history of active suicidal ideation * Alcohol/drug problems in the past year or lifetime alcohol or drug dependence * Medications that act on the central nervous system * History of seizure * History of epilepsy * Increased risk of seizure for any reason * Pregnancy, or positive pregnancy test * IQ \<80 * Any medical condition that increases risk for fMRI or TMS * Any metal in their body which would make having an MRI scan unsafe * Any sort of medical implants * Hearing loss * Claustrophobia
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Anxiety Potentiated Startle | Pre and 24-hours post stimulation | Electromyography Facial electromyography (EMG) startle responses were recorded from the left orbicularis oculi muscle at 2000 Hz using a Biopac MP160 unit (Biopac; Goleta, CA) via 15 × 20 mm hydrogel coated vinyl electrodes (Rhythmlink #DECUS10026; Columbia, SC). Startle EMG was bandpass filtered from 30 to 300 Hz, rectified, and smoothed using a 20-ms sliding window. Startle responses were scored as the peak (max during the 20 ms to 120 ms post-noise window) - the baseline (50 ms pre-noise window), and converted to t-scores with a mean of 50 and a standard deviation of 10 (tx = \[Zx × 10\] + 50). Greater t-scores mean larger blinks, which could be associated with greater anxiety, however there is no clinically relevent threshold. Noisy trials (baseline SD \> 2x run SD) were excluded, and no blink (peak \< baseline range) trials were coded as 0. To calculate APS, we subtracted the response during the neutral ITI from the response during the unpredictable ITI. |
| Fear Potentiated Startle | Pre and 24 hours post stimulation | Electromyography Facial electromyography (EMG) startle responses were recorded from the left orbicularis oculi muscle at 2000 Hz using a Biopac MP160 unit (Biopac; Goleta, CA) via 15 × 20 mm hydrogel coated vinyl electrodes (Rhythmlink #DECUS10026; Columbia, SC). Startle EMG was bandpass filtered from 30 to 300 Hz, rectified, and smoothed using a 20-ms sliding window. Startle responses were scored as the peak (max during the 20 ms to 120 ms post-noise window) - the baseline (50 ms pre-noise window), and converted to t-scores with a mean of 50 and a standard deviation of 10 (tx = \[Zx × 10\] + 50). Greater t-scores mean larger blinks, which could be associated with greater fear, however there is no clinically relevent threshold. Noisy trials (baseline SD \> 2x run SD) were excluded, and no blink (peak \< baseline range) trials were coded as 0. To calculate FPS, we subtracted the response during the predictable ITI from the response during the predictable Cue. |
| Sternberg WM Accuracy | Pre and 24 hours post stimulation | Sternberg task: On each WM trial, subjects will see a series of 4 letters presented singularly (encoding period) that will be followed by a brief interval where subjects are required to maintain these letters (maintenance period). At the end of the maintenance period, subjects will be prompted to make a response based on the task instructions (response period). The response prompt will consist of a letter and a number. The letter will be chosen from the study series, and the number will correspond to a position in the series. The subjects will indicate whether the position of the letter in the series matches the number. |
| TMS-evoked BOLD Responses | Responses are measured within the TMS/fMRI session in response to each TMS pulse and collapsed across trials. There is no sham condition. This session was typically conducted during the washout period, but varied depending upon participant schedule. | As with Experiment 1, subjects will have Neutral, Predictable, and Unpredictable periods. During the neutral periods, they will be safe from shocks. During the predictable periods, they can receive shocks but only when there is a cue present. During the unpredictable periods, they are at risk for shock during the entire duration of the block. Rather than probing their ongoing anxiety with the startle probes, we replaced the startle probes with single TMS pulses to the right dlPFC. This allowed us to causally examine the effect of right dlPFC activity (induced by the TMS pulse) on the neural activity. BOLD responses are collapsed across regions and conditions to examine right dlPFC BOLD down regulation. |
Countries
United States
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| cTBS Arm Active stimulation. A Magventure MagPro 100X stimulator with a B65AP (active/placebo) figure-8 coil was used for the cTBS sessions. The active and sham coil sides of the coil were masked and assigned blinded labels (e.g. A = active, B = sham). The label key was maintained by a member of the study staff not directly involved in the collection or analysis of the data. All other study staff were blinded to the label assignments.
cTBS parameters. During each cTBS session, a single 600 pulse cTBS train was delivered during each stimulation session at 100% of RMT. The train consisted of 50 Hz bursts, repeated at intervals of 200 ms (5 Hz) for 40 sec. | 28 |
| iTBS Arm Active stimulation. A Magventure MagPro 100X stimulator with a B65AP (active/placebo) figure-8 coil was used for the cTBS sessions. The active and sham coil sides of the coil were masked and assigned blinded labels (e.g. A = active, B = sham). The label key was maintained by a member of the study staff not directly involved in the collection or analysis of the data. All other study staff were blinded to the label assignments.
iTBS parameters. During each iTBS session, ten 60-pulse iTBS trains were delivered at 100% of RMT. The trains consisted of 3 50 Hz bursts repeated at intervals of 200 ms (5 Hz) for 2 seconds of stimulation separated by 8 second gaps, for a total of 600 pulses per session. | 28 |
| Total | 56 |
Baseline characteristics
| Characteristic | cTBS Arm | iTBS Arm | Total |
|---|---|---|---|
| Age, Continuous | 26.61 years STANDARD_DEVIATION 7.04 | 23.75 years STANDARD_DEVIATION 4.26 | 25.18 years STANDARD_DEVIATION 5.65 |
| Ethnicity (NIH/OMB) Hispanic or Latino | 2 Participants | 2 Participants | 4 Participants |
| Ethnicity (NIH/OMB) Not Hispanic or Latino | 26 Participants | 26 Participants | 52 Participants |
| Ethnicity (NIH/OMB) Unknown or Not Reported | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) American Indian or Alaska Native | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Asian | 4 Participants | 6 Participants | 10 Participants |
| Race (NIH/OMB) Black or African American | 7 Participants | 4 Participants | 11 Participants |
| Race (NIH/OMB) More than one race | 0 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) Native Hawaiian or Other Pacific Islander | 0 Participants | 0 Participants | 0 Participants |
| Race (NIH/OMB) Unknown or Not Reported | 0 Participants | 1 Participants | 1 Participants |
| Race (NIH/OMB) White | 17 Participants | 16 Participants | 33 Participants |
| Region of Enrollment United States | 28 participants | 28 participants | 56 participants |
| Sex: Female, Male Female | 21 Participants | 18 Participants | 39 Participants |
| Sex: Female, Male Male | 7 Participants | 10 Participants | 17 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk | EG003 affected / at risk |
|---|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 34 | 0 / 34 | 0 / 34 | 0 / 34 |
| other Total, other adverse events | 3 / 34 | 3 / 34 | 0 / 34 | 0 / 34 |
| serious Total, serious adverse events | 0 / 34 | 0 / 34 | 0 / 34 | 0 / 34 |
Outcome results
Primary
Anxiety Potentiated Startle
Electromyography Facial electromyography (EMG) startle responses were recorded from the left orbicularis oculi muscle at 2000 Hz using a Biopac MP160 unit (Biopac; Goleta, CA) via 15 × 20 mm hydrogel coated vinyl electrodes (Rhythmlink #DECUS10026; Columbia, SC). Startle EMG was bandpass filtered from 30 to 300 Hz, rectified, and smoothed using a 20-ms sliding window. Startle responses were scored as the peak (max during the 20 ms to 120 ms post-noise window) - the baseline (50 ms pre-noise window), and converted to t-scores with a mean of 50 and a standard deviation of 10 (tx = \[Zx × 10\] + 50). Greater t-scores mean larger blinks, which could be associated with greater anxiety, however there is no clinically relevent threshold. Noisy trials (baseline SD \> 2x run SD) were excluded, and no blink (peak \< baseline range) trials were coded as 0. To calculate APS, we subtracted the response during the neutral ITI from the response during the unpredictable ITI.
Time frame: Pre and 24-hours post stimulation
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| cTBS Arm (Active/Sham) | Anxiety Potentiated Startle | active | 4.600026484 T-scores | Standard Deviation 3.690076704 |
| cTBS Arm (Active/Sham) | Anxiety Potentiated Startle | sham | 2.098026426 T-scores | Standard Deviation 3.687256948 |
| iTBS Arm (Active/Sham) | Anxiety Potentiated Startle | sham | 1.347655375 T-scores | Standard Deviation 5.091633437 |
| iTBS Arm (Active/Sham) | Anxiety Potentiated Startle | active | 3.411297242 T-scores | Standard Deviation 4.494552916 |
| cTBS Arm (Sham/Active) | Anxiety Potentiated Startle | active | 3.122643391 T-scores | Standard Deviation 4.286242167 |
| cTBS Arm (Sham/Active) | Anxiety Potentiated Startle | sham | 3.063363512 T-scores | Standard Deviation 4.002345814 |
| iTBS Arm (Sham/Active) | Anxiety Potentiated Startle | active | 4.848463905 T-scores | Standard Deviation 2.77573715 |
| iTBS Arm (Sham/Active) | Anxiety Potentiated Startle | sham | 3.551809142 T-scores | Standard Deviation 4.504709597 |
Primary
Fear Potentiated Startle
Electromyography Facial electromyography (EMG) startle responses were recorded from the left orbicularis oculi muscle at 2000 Hz using a Biopac MP160 unit (Biopac; Goleta, CA) via 15 × 20 mm hydrogel coated vinyl electrodes (Rhythmlink #DECUS10026; Columbia, SC). Startle EMG was bandpass filtered from 30 to 300 Hz, rectified, and smoothed using a 20-ms sliding window. Startle responses were scored as the peak (max during the 20 ms to 120 ms post-noise window) - the baseline (50 ms pre-noise window), and converted to t-scores with a mean of 50 and a standard deviation of 10 (tx = \[Zx × 10\] + 50). Greater t-scores mean larger blinks, which could be associated with greater fear, however there is no clinically relevent threshold. Noisy trials (baseline SD \> 2x run SD) were excluded, and no blink (peak \< baseline range) trials were coded as 0. To calculate FPS, we subtracted the response during the predictable ITI from the response during the predictable Cue.
Time frame: Pre and 24 hours post stimulation
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| cTBS Arm (Active/Sham) | Fear Potentiated Startle | active | 7.598819444 T-scores | Standard Error 4.007589756 |
| cTBS Arm (Active/Sham) | Fear Potentiated Startle | sham | 5.61775824 T-scores | Standard Error 3.484582125 |
| iTBS Arm (Active/Sham) | Fear Potentiated Startle | sham | 3.016188249 T-scores | Standard Error 6.334558956 |
| iTBS Arm (Active/Sham) | Fear Potentiated Startle | active | 6.719283178 T-scores | Standard Error 5.508434373 |
| cTBS Arm (Sham/Active) | Fear Potentiated Startle | sham | 6.651174586 T-scores | Standard Error 5.405511643 |
| cTBS Arm (Sham/Active) | Fear Potentiated Startle | active | 6.564791401 T-scores | Standard Error 4.771055896 |
| iTBS Arm (Sham/Active) | Fear Potentiated Startle | sham | 6.496396165 T-scores | Standard Error 3.242895129 |
| iTBS Arm (Sham/Active) | Fear Potentiated Startle | active | 5.529855023 T-scores | Standard Error 4.175930826 |
Primary
Sternberg WM Accuracy
Sternberg task: On each WM trial, subjects will see a series of 4 letters presented singularly (encoding period) that will be followed by a brief interval where subjects are required to maintain these letters (maintenance period). At the end of the maintenance period, subjects will be prompted to make a response based on the task instructions (response period). The response prompt will consist of a letter and a number. The letter will be chosen from the study series, and the number will correspond to a position in the series. The subjects will indicate whether the position of the letter in the series matches the number.
Time frame: Pre and 24 hours post stimulation
| Arm | Measure | Group | Value (MEAN) | Dispersion |
|---|---|---|---|---|
| cTBS Arm (Active/Sham) | Sternberg WM Accuracy | active | 80.70887446 percent correct | Standard Deviation 15.81409388 |
| cTBS Arm (Active/Sham) | Sternberg WM Accuracy | sham | 81.38888889 percent correct | Standard Deviation 16.70756441 |
| iTBS Arm (Active/Sham) | Sternberg WM Accuracy | sham | 87.85342262 percent correct | Standard Deviation 7.303273682 |
| iTBS Arm (Active/Sham) | Sternberg WM Accuracy | active | 86.99776786 percent correct | Standard Deviation 10.46204331 |
| cTBS Arm (Sham/Active) | Sternberg WM Accuracy | active | 84.89010989 percent correct | Standard Deviation 13.40836505 |
| cTBS Arm (Sham/Active) | Sternberg WM Accuracy | sham | 86.76739927 percent correct | Standard Deviation 16.03952935 |
| iTBS Arm (Sham/Active) | Sternberg WM Accuracy | active | 87.45039683 percent correct | Standard Deviation 10.42233406 |
| iTBS Arm (Sham/Active) | Sternberg WM Accuracy | sham | 90.35218254 percent correct | Standard Deviation 8.095574634 |
Primary
TMS-evoked BOLD Responses
As with Experiment 1, subjects will have Neutral, Predictable, and Unpredictable periods. During the neutral periods, they will be safe from shocks. During the predictable periods, they can receive shocks but only when there is a cue present. During the unpredictable periods, they are at risk for shock during the entire duration of the block. Rather than probing their ongoing anxiety with the startle probes, we replaced the startle probes with single TMS pulses to the right dlPFC. This allowed us to causally examine the effect of right dlPFC activity (induced by the TMS pulse) on the neural activity. BOLD responses are collapsed across regions and conditions to examine right dlPFC BOLD down regulation.
Time frame: Responses are measured within the TMS/fMRI session in response to each TMS pulse and collapsed across trials. There is no sham condition. This session was typically conducted during the washout period, but varied depending upon participant schedule.
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| cTBS Arm (Active/Sham) | TMS-evoked BOLD Responses | -0.0074 BOLD signal | Standard Deviation 0.0056 |
| iTBS Arm (Active/Sham) | TMS-evoked BOLD Responses | -0.0063 BOLD signal | Standard Deviation 0.0072 |