Accelerated Theta Burst Stimulation (aTBS): Safety, efficacy and future advancements

Abstract

Theta-burst stimulation (TBS) is a non-invasive brain stimulation technique that can be used to modulate neural networks underlying psychiatric and neurological disorders. TBS can be delivered intermittently (iTBS) or continuously (cTBS). The conventional iTBS protocol is FDA-approved to treat otherwise treatment-resistant depression but the 6-week duration limits the applicability of this therapy. Accelerated TBS protocols present an opportunity to deliver higher pulse-doses in shorter periods of time, resulting in faster and potentially more clinically effective treatment. However, the acceleration of TBS delivery raises questions regarding the relative safety, efficacy and durability compared to conventional TBS protocols. In this review paper we present the data from accelerated TBS trials to date which support that accelerated protocols appear safe and effective, while acknowledging the need for more durability data. We discuss the stimulation parameters that seem to be important for the efficacy of accelerated TBS protocols and possible avenues for further optimization.

Introduction

Theta burst stimulation (TBS) is a form of repetitive transcranial magnetic stimulation (rTMS), a non-invasive brain stimulation method used to modulate neural networks underlying psychiatric and neurological disorders (1–3). rTMS protocols have been FDA-approved to treat treatment-resistant depression (TRD), obsessive-compulsive disorder (OCD), smoking cessation, migraine and anxious depression. Initial rTMS protocols involved the delivery of magnetic pulses at equally spaced intervals at either high (~10Hz) or low (~1Hz) frequency with the goal of increasing or decreasing activity in the target neural network respectively. In 2005, TBS was developed as a more efficient form of rTMS (3). The stimulation pattern of TBS mimics the natural activity of the brain during learning tasks, with three biphasic pulses delivered every 20 ms (50 Hz) in bursts separated by 200 ms (5 Hz). TBS can be delivered continuously (cTBS) which is considered inhibitory or delivered intermittently (iTBS) in 2 second trains separated by 8 second intervals, which is considered excitatory (3). The excitatory and inhibitory effect of iTBS and cTBS respectively is, however, a simplistic view as studies have shown a wide range of individual variability in response to these protocols (4–6). Furthermore, it has been shown that varying protocol parameters such as pulse number may reverse the effect of TBS protocols (7,8).

TBS protocols have advantages over conventional rTMS protocols due to the shorter duration of stimulation sessions and apparent efficacy at lower stimulation intensities whilst exhibiting equivalent safety and clinical efficacy (3,9,10). A large, multicenter clinical trial found iTBS had equivalent antidepressant efficacy to 10Hz rTMS and consequently iTBS was FDA-approved for the treatment of TRD (9). Of note, FDA-approved conventional 10Hz and iTBS protocols both deliver stimulation at an intensity of 120% resting motor threshold (rMT). Currently, no cTBS protocols are FDA-approved and iTBS is not FDA-approved to treat other conditions. Despite FDA-approval of iTBS for TRD, clinical utility is limited by the 6-week duration and remission rate of 32% (9). Rapid-acting, effective treatments are needed particularly for suicidal and inpatient patient populations for which urgency is greatest. Accelerated TBS (aTBS) protocols, consisting of multiple TBS sessions per day, could meet this clinical need by delivering higher pulse-doses in shorter periods of time. Accelerated protocols could also allow more patients to be treated in the same time period and reduce the number of days patients are required to travel to TMS clinics. However, accelerating the delivery of TBS sessions raises questions regarding the relative safety, efficacy and durability compared to the classic 6-week iTBS protocol. In this paper we review the existing aTBS data from both TRD patients and other patient populations to discuss the safety and efficacy of aTBS, the limitations of the current data and avenues for future advancement of aTBS.

Safety of aTBS

To date, there are reports of 943 individuals who have received active aTBS across 33 different studies (11–17). See Table 1 and 2. The total number of sessions delivered in an aTBS course have ranged from 3 to 104 with the number of daily sessions ranging from 2 to 10. No serious adverse events have been reported for any aTBS protocols to date. Of the 33 publications, 22 reported side-effect data. The most common reported side-effect for aTBS has been headache with an incidence of 31.3%. This is consistent with non-accelerated TBS/rTMS protocols, for which headache has been reported as the most common side-effect. Other relatively commonly reported aTBS side-effects include fatigue (7.6%) and nausea (5%). See Table 1 and 2 for detailed information on reported side-effects.

Table 1.

Accelerated theta-burst stimulation studies in MDD.

Authors	Number of Overall Participants	Number of Accelerated TMS Participants	Number of Other Condition Participants	Diagnosis	Stimulation Location	Targeting method	TMS Intensity (% Motor Threshold)	TMS Frequency (Hz)	Total Pulses Per Session	Number of Sessions Per Day	Intersession Interval (Minutes)	Number of Treatment Days	Total Number of Sessions	Outcomes on Primary Results	Response	Reported Side Effects
Baeken et al. (20)	1	1	0	Treatment Resistant Depression	Left DLPFC and Right DLPFC	Optical Neuronavigation	80% rMT	iTBS and cTBS	600 pulses/session for both iTBS and cTBS	5 sessions/day for both iTBS and cTBS	15 minutes	4 days of cTBS preceding 4 days of iTBS	40 total sessions: 20 accelerated iTBS and 20 accelerated cTBS	HDRS-17 symptom reduction, relapse at 2 months post-TMS assessment	/	headache, fatigue, scalp sensitivity: 1
Blumberger et al. (18)	208	103	105 once daily iTBS	Major Depressive Disorder	Left DLPFC	Optical Neuronavigation	120% rMT	iTBS	600 pulses/session	2 sessions/day	60 minutes	30 days	60 total sessions	Twice daily group: 44.3% response rate (22.7% remission); Once daily group: 41.1% response rate (23.1% remission) after 30 days of treatment; significant iTBS effect between baseline and post-treatment in both groups; no significant difference between groups	44.3	Twice daily group: headache: 72, unrelated medical condition: 53, nausea: 23, dizziness: 19, fatigue: 19, insomnia: 18, facial pain: 15, anxiety: 11, back/neck pain: 8, facial twitching: 7, migraine: 7, abnormal sensation: 7, vivid dreams: 4, unrelated accident: 4, vomiting: 4, blurred vision: 4, tinnitus: 2, confusion: 1
Bröcker et al. (26)	9	9	0	Major Depressive Disorder or Bipolar Depression	Left DLPFC	5.5 cm anterior of motor site	80% rMT	iTBS	1782 pulses/session	2.5 sessions/day	20 minutes	8 days	20 total sessions	CES-D symptom reduction (55.5% response); no remission rates reported	55.5	Not reported
Cantù et al. (27)	6	6	0	Major Depressive Disorder or Bipolar Depression	Left DLPFC	5 cm anterior of motor site	80% rMT	30Hz iTBS	600 pulses/session	3 sessions/day	15 minutes	6 days	18 total sessions	neither response nor remission as evaluated with HDRS-17	0	Not reported
Chen et al. (29)	295	211	84 once daily 10Hz	Major Depressive Disorder or Bipolar Disorder	Left DLPFC and Right DLPFC	Beam F3	80% (cTBS/iTBS) or 120% (10Hz or cTBS/iTBS) rMT	Sequential cTBS/iTBS or 10Hz (active control)	10Hz: 3000 pulses/session; cTBS/iTBS: 600 pulses/session each	10Hz: 1; cTBS/iTBS: 2 x 2–3	15 minutes	10Hz: 20 days; cTBS/iTBS: 8 days	10Hz: 20 total sessions; cTBS/iTBS: 40 total sessions	10Hz condition: response rate (51.4%), remission rate (34.7%); cTBS/iTBS 80% rMT condition: response rate (44.1%), remission rate (26.9%); cTBS/iTBS 120% rMT condition: 36.8%, remission: 24.1%. No significant differences between conditions	44.1	Not reported
Cole et al. (21)	21	21	0	Treatment Resistant Depression	Left DLPFC	Optical Neuronavigation	90% rMT	iTBS	1800 pulses/session	10 sessions/day	50 minutes	5 days	50 total sessions	MADRS symptom reduction (90.5% response, 90.5% remission)	90.5	No side-effects
Cole et al. (22)	29	15	14 sham	Treatment Resistant Depression	Left DLPFC	Optical Neuronavigation	90% rMT depth corrected	iTBS	1800 pulses/session	10 sessions/day	50 minutes	5 days	50 total sessions	active group:85.7% response/78.6%; sham group: 26.7% response/13.3% remission	85.7	fatigue: 8, neck/back discomfort: 7, headache: 8, discomfort at treatment site: 5, anxiety: 4, jaw discomfort: 2, dental issues: 1
Duprat et al. (23)	50	50	25 sham first	Treatment Resistant Depression	Left DLPFC	Optical Neuronavigation	110% rMT	iTBS	1620 pulses/session	5 sessions/day	15 minutes	4 days	20 total sessions	HDRS-24 symptom reduction immediately after end of treatment (28% response/5% remission ); 2 weeks post-TMS (38% response/30% remission);	38	headache: 20, scalp pain: 4, fatigue: 4, disorientation: 2, stress: 1, transient worsening of depression: 1, difficulty focusing: 1
Fitzgerald et al. (19)	74	36	38 once daily TMS	Major Depressive Disorder	Left DLPFC	Beam F3	120% rMT	iTBS	600 pulses/session	3 sessions/day	15 minutes	7 days	21 total sessions	MADRS symprom reduction (19.4% response/8.3% remission) by end of treatment;4 weeks post-treatment: 27.8% response/8.3% response	27.8	headache: 12
George et al. (30)	1	1	0	Treatment Resistant Depression	Left DLPFC	Beam F3 (F3, 10–20 international system)	120% rMT	iTBS	600 pulses/session	7 sessions/day	30 minutes	3 days	21 total sessions	PHQ-9 response	/	Not reported
Konstantinou et al. (24)	1	1	0	Treatment Resistant Depression	Left DLPFC	Beam F3	110% rMT	iTBS	1st course: 1800 pulses/session; 2nd course: 600 pulses/session	8 sessions/day for both treatment courses	50 minutes	5 days for both treatment courses	1st course: 96 total sessions: 40 aTMS and 64 of 8x daily tapering; 2nd course: 104 total sessions: 40 aTMS and 64 of 8x daily tapering	PHQ-9 and BDI response and remission,relapse, response and remission after second treatment round	/	No side-effects
Konstantinou et al. (25)	1	1	0	Treatment Resistant Depression	Left DLPFC	Optical Neuronavigation	120% rMT	iTBS	1800 pulses/session	8 sessions/day	50 minutes	5 days	1st course: 104 total sessions: 40 aTMS and 64 of 8x daily tapering; 2nd course: 104 total sessions: 40 aTMS and 64 of 8x daily tapering	BDI and PHQ-9 response and complete remission 2 months after treatment	/	No side-effects
Mielacher et al. (13)	53	35	18 sham	Major depressive disorder (MDD	Left DLPFC and parietal targets functionally connected to the hippocampus.	Optical Neuronavigation	80% rMT	iTBS	600 pulses/session	2 sessions per day for LDLPFC only group or 3 sessions/day for LDLPFC + bilaterial iLPC group	5 minutes	15 days	30 total sessions for LDLPFC only group, 45 total sessions for LDLPFC + bilateral iLPC group	Symptom reduction for all conditions. LDLPFC only group showed best response.	/	13 headache, 4 nausea, 9 dizziness, 21 muscle twitching, 14 pain
Stöhrmann et al.. (16)	20	15	5	Treatment-resistant depression	ft and right DLPF	Optical Neuronavigation	120% rMT	iTBS and cTBS	600 pulses/session	2 sessions/day	60 minutes	15 days	30 total sessions	atients responded, 3 went into remiss	53	Not reported
Struckmann et al. (47)	51	25	26 sham	Major Depressive Disorder or Schizophrenia	DMPFC	Optical Neuronavigation	90% foot rMT	iTBS	600 pulses/session	2 sessions/day	15 minutes	10 days	20 total sessions	No improvement above placebo in cognitive task	/	Not reported
Williams et al. (28)	6	6	0	Major Depressive Disorder or Bipolar Depression	Left DLPFC	Optical Neuronavigation	120% rMT	iTBS	1800 pulses/session	10 sessions/day	50 minutes	5 days	50 total sessions	HDRS-17 symptom reduction (83.3% response, 66.7% remission)	83.3	Not reported

Table 2.

Accelerated theta-burst stimulation studies in patient populations other than MDD.

Authors	Number of Overall Participants	Number of Accelerated TMS Participants	Number of Other Condition Participants	Diagnosis	Stimulation Location	Targeting method	TMS Intensity (% Motor Threshold)	TMS Frequency (Hz)	Total Pulses Per Session	Number of Sessions Per Day	Intersession Interval (Minutes)	Number of Treatment Days	Total Number of Sessions	Outcomes on Primary Results	Response	Reported Side Effects
Zhao et al. (58)	83	83	74 completers: 26 iTBS over left DLPFC, 18 cTBS over left DLPFC, and 30 cTBS over right DLPFC	Abstinent Methamphetamine Dependence	Left DLPFC or Right DLPFC	Optical Neuronavigation	70% rMT	iTBS or cTBS	600 pulses/session for both iTBS and cTBS	2 sessions/day for both iTBS and cTBS	240 minutes	5 days	104	LDLPFC iTBS and RDLPFC cTBS Cued-Craving scale symptom reduction; no such effect in LDLPFC cTBS	16.67% response for cTBS LDLPFC, 36.67% response for cTBS RDLPFC and 23.08% response for iTBS LDLPFC	Side effects not reported per participant
Wu et al. (57)	49	25	24 sham	Alzheimer’s Disease	Left DLPFC	Optical Neuronavigation	70% rMT	iTBS	600 pulses/session	3 sessions/day	15 minutes	14 days	42 total sessions	significant improvement in associative memory for active (72.8% normalized change) but not sham (-10.5% normalized change). Effect stronger for higher baseline MMSE scores lasting up to 8 weeks.	45.83% response in the active group, 8.69% in the sham group	scalp pain: 3, eyelid twitches: 2
Bulteau et al. (104)	26	12	14 sham	Bipolar Depression	Left DLPFC	Optical Neuronavigation	80% rMT	iTBS	990 pulses/session	2 sessions/day	180 minutes	15 days	30 total sessions	MADRS symptom reduction (75% response), 42% remission (active TMS)	75	No side-effects
Bröcker et al. (26)*	9	9	0	Bipolar Depression or Major Depressive Disorder	Left DLPFC	5.5 cm anterior of motor site	80% rMT	iTBS	1782 pulses/session	2.5 sessions/day	20 minutes	8 days	20 total sessions	CES-D symptom reduction (55.5% response); no remission rates reported	55.5	Not reported
Cantù et al. (27)*	6	6	0	Bipolar Depression or Major Depressive Disorder	Left DLPFC	5 cm anterior of motor site	80% rMT	30Hz iTBS	600 pulses/session	3 sessions/day	15 minutes	6 days	18 total sessions	neither response nor remission as evaluated with HDRS-17	0	Not reported
Williams et al. (28)*	6	6	0	Bipolar Depression or Major Depressive Disorder	Left DLPFC	Optical Neuronavigation	120% rMT	iTBS	1800 pulses/session	10 sessions/day	50 minutes	5 days	50 total sessions	HDRS-17 symptom reduction (83.3% response, 66.7% remission)	83.3	Not reported
Chen et al. (29)*	295	211	84 once daily 10Hz	Bipolar Disorder or Major Depressive Disorder	Left DLPFC and Right DLPFC	Beam F3	80% (cTBS/iTBS) or 120% (10Hz or cTBS/iTBS) rMT	Sequential cTBS/iTBS or 10Hz (active control)	10Hz: 3000 pulses/session; cTBS/iTBS: 600 pulses/session each	10Hz: 1; cTBS/iTBS: 2 x 2–3	15 minutes	10Hz: 20 days; cTBS/iTBS: 8 days	10Hz: 20 total sessions; cTBS/iTBS: 40 total sessions	10Hz condition: response rate (51.4%), remission rate (34.7%); cTBS/iTBS 80% rMT condition: response rate (44.1%), remission rate (26.9%); cTBS/iTBS 120% rMT condition: 36.8%, remission: 24.1%. No significant differences between conditions	44.1	Not reported
Steele et al. (14)	9	9	/	Cocaine use disorder	Left DLPFC	Beam F3	100% rMT	iTBS	600 pulses/sessio	3 sessions/day	60 minutes	10 days	30 total sessions	78% cocaine reduction ($ spent p/w) and 70% reduction in days of use. Consumption of nicotine, alcohol, and THC reduced.	/	headache: 9, eye pain: 1, intermittent tinnitus: 1, muscle soreness in right arm: 1, right-hand supination/pronation at the wrist: 1, mild dizziness: 1, sudden waking: 1
Watt et al. (59)	2	2	0	Existential Distress	Left DLPFC	Beam F3	80% rMT	iTBS	600 pulses/session	8 sessions/day	45 minutes	5 days	40 total sessions	response and remission in both patients	100	No side-effects
Wu et al. (53)	60	20	20 sham, 20 once daily 20Hz	Healthy Adults	Left DLPFC	Optical Neuronavigation	70% rMT	iTBS	600 pulses/session	3 sessions/day	15 minutes	1 day	3 total sessions	iTBS and 20Hz conditions: increased accuracy in 3-back task; no improvement during sham	/	Not reported
Yu et al. (54)	16	16	0	Healthy Adults	Left Primary Motor Cortex	based on MEPs from non-dominant FDI	70% rMT	iTBS	600 pulses/session	3 sessions/day	Variable within-subjects: 0, 10, or 30 minutes	3 days	9 total sessions	MEPs affected by stimulation pause lenghth, with 30 minutes > 10 minutes > 0 minutes	/	No side-effects
Dutta et al. (50)	33	18	15 sham	Obsessive Compulsive Disorder	Orbitofrontal Cortex	10% dorsal from the nasion, 10% lateral (Fp1, 10–20 international system))	80% rMT	cTBS	600 pulses/session	2 sessions/day	30 minutes	5 days	10 total sessions	cTBS significnat OCD symptom reduction for active;effects did not survive correctiong for moderators	/	headache: 3
Mukherjee et al. (17)	26	13	13 sham	Obsessive Compulsive Disorder	SMA	Optical Neuronavigation	120% rMT	iTBS and cTBS	600 pulses per hemisphere	2 sessions per day	1 hour	15 days	30 total sessions	YBOCS total,obsession, and compulsion sub scores significantly decreased in active TBS group compared to sham	3 responders on the YBOCS in the active group (23.1%) and 0% in sham group.	headache: 3
Williams et al. (49)	7	7	0	Obsessive Compulsive Disorder	Right Frontal Pole	Optical Neuronavigation	90% rMT depth corrected	cTBS	1800 pulses/session	10 sessions/day	60 minutes	5 days	50 total sessions	response: 4/7 patients, lasting effect 28 days post-treatment (3/7 repsonders)	57.14	headache: 4, fatigue: 3
Ji et al. (52)	42	22	20 sham	Parkinson's Disease	Left SMA	Optical Neuronavigation	80% rMT	cTBS	600 pulses/session	3 sessions/day	15 minutes	14 days	42 total sessions	Active protocol: significant PD symptom reduction	/	No side-effects
Trung et al. (51)	28	14	14 sham	Parkinson's Disease	Left DLPFC	Optical Neuronavigation	80% rMT	iTBS	600 pulses/session	2 sessions/day	60 minutes	3 days	6 total sessions	No improvement above placebo in cognitive task neuropsychological assessment	/	No side-effects
Zhang et al. (56)	42	28	14 sham	Post-Stroke Hemiparesis	Ipsilesional Primary Motor Cortex	Optical Neuronavigation	70% rMT or 20% rMT	Sequential cTBS and iTBS	600 pulses/session for both iTBS and cTBS	2 sessions/day (1 iTBS, 1 cTBS)	10 minutes	10 days	20 total sessions: 10 accelerated iTBS and 10 accelerated cTBS	Priming 70% rMT cTBS prior to 70% rMT iTBS caused the most significant motor improvements, followed by 20% rTMS cTBS prior to 70% rMT iTBS, and sham	/	Not reported
Zhang et al. (55)	62	31	31 healthy controls for neuroimaging comparison	Postpartum Depression	Left DLPFC	Optical Neuronavigation	iTBS	1800 pulses/session	10 sessions/day	50 minutes	5 days	50 total sessions	HDRS-17 significant symptom reduction;	/	/	Not reported
Chauhan et al. (48)	35	18	17 sham	Schizophrenia (Treatment Resistant)	Midline Cerebellum	1 cm below the inion (Iz, 10–20 international system)	80% rMT	iTBS	600 pulses/session	2 sessions/day	30 or more minutes	5 days	10 total sessions	No significant above sham effect on schizophrenia symptoms	/	headache: 5
Jin et al. (12)	60	30	30 sham	Schizophrenia	LDLPFC	Optical Neuronavigation	120% rMT	iTBS	600 pulses/session	3 sessions/day	15 minutes	20 days	60 sessions	FERT scores for active iTBS group differed significantly from sham iTBS group 2 weeks and 4 weeks of treatment. PANSS score and negative symptom score differed significantly at week 4.	/	headache: 30%, numb headedness: 33%, needle-tingling sensation: 20%, dizziness: 17%
Struckmann et al. (47)*	51	25	26 sham	Schizophrenia or Major Depressive Disorder	DMPFC	Optical Neuronavigation	iTBS	600 pulses/session	2 sessions/day	15 minutes	10 days	20 total sessions	No improvement above placebo in cognitive task	/	/	Not reported
Mikellides et al. (15)	89	59	30	Smoking cessation	Left DLPFC	Beam F3	100% rMT	iTBS	600 pulses/session	4 sessions/day	30 minutes	5 days	20 total sessions	iTBS reduced cigarette consumption equally for real and sham	/	headache: 6, sleepiness: 3, insomnia: 1, tension: 1, numbness on stimulation site: 1, lightheadedness: 1, coughiness: 1, numbness on stim site & sleepiness: 1, mild headache & sleepiness: 1

^*Indicates studies also listed in Table 1

It should be noted that the adverse event data reported above include all aTBS protocols across different stimulation sites and patient populations. In order to compare adverse event profiles across accelerated and non-accelerated TBS protocols, the adverse events for aTBS studies that delivered left-dorsolateral prefrontal cortex (LDLPFC) stimulation in patients with Major Depressive Disorder (MDD) are reported in comparison to the adverse events in the pivotal THREE-D trial that led to the Food and Drug Administration (FDA) approval of conventional TBS. To date, there have been 16 aTBS trials in MDD participants, 15 of these stimulated LDLPFC (see Table 1) but only 9 of these studies reported side-effect data (13,18–25). The most common side-effect for LDLPFC aTBS studies for MDD (9 studies, 263 participants) was headache (48%) which occurred at a similar rate to that reported for conventional once daily iTBS (65%). Fatigue (12.7%) and nausea (10.3%) also occurred at similar rates to those reported in the THREE-D trial for iTBS (fatigue 8% and nausea 7%). It should be noted that two of these studies stimulated other brain areas (right DLPFC and parietal targets) in addition to LDLPFC (see Table 1) and other stimulation parameters such as pulse number, stimulation intensity and targeting method were not consistent with conventional iTBS.

The dropout rate for aTBS studies to date is lower than is reported for conventional iTBS. The dropout rate across aTBS studies to date is 3.3% and 1.6% for LDLPFC aTBS studies in MDD patients. In comparison, the dropout rate in the THREE-D trial was 8% for conventional iTBS (9). The lower dropout rate for aTBS than for conventional iTBS could reflect preferable treatment schedules compared to the 6-week conventional iTBS schedule or the lower stimulation intensities often used in aTBS studies. However, given the limited literature, further studies are required to determine whether there are significant differences in tolerability between accelerated and non-accelerated protocols.Trials directly comparing accelerated and conventional TBS protocols with all other stimulation parameters kept constant are required. None of the aTBS studies to date have delivered a protocol with the same number of sessions, pulses per session, overall pulse-dose and stimulation intensity as the conventional TBS protocol. See Table 1 and Table 2 for stimulation parameters used in aTBS studies.

It is possible that differences in side-effect data collection methods between aTBS studies and the THREE-D trial may have biased the data. The methods by which adverse event data were collected were not consistent across aTBS studies i.e. structured assessments (13,20), unstructured questioning (18,26) or relying on spontaneous reporting or observation (19,21–23). In addition, long-term side-effect data for aTBS is not available due to the short follow-up time periods in the existing literature (13,16,18–30). It is possible that long-term or delayed onset side-effects exist, although this is not expected due to the minimal side-effect profile of conventional TBS/rTMS (9,31–34).

Clinical efficacy of aTBS for depression

There have been 16 reports of aTBS trials for individuals with MDD to date (13,16,18–30,47) and 15 of these stimulated the LDLPFC. See Table 1. Ten of these studies had more than 1 participant and reported response rates for depression symptom reduction. Nine of these studies reported remission rates. The average response and remission rates from these studies were 52.2% (n=10) and 38.2% (n=9) respectively. These response and remission rates are similar to those reported for iTBS in the THREE-D trial (49% 32%(9)). However, multiple stimulation parameters beyond the accelerated delivery, differed from conventional TBS and were inconsistent across studies. The overall pulse dose, number of sessions, stimulation intensity, targeting method and pulses per session varied across aTBS studies (see Table 1). In addition, two of these studies stimulated the right DLPFC as well as the LDLPFC. The inconsistency in stimulation parameters between conventional iTBS and aTBS protocols makes it challenging to compare clinical efficacy. A direct comparison of accelerated and conventional rTMS has been previously conducted (35). In this study, the accelerated form of rTMS involved three sessions of rTMS delivered per day across three weeks with 1–3 stimulation days per week for a total of 18 sessions. This was compared to a course of 20 daily stimulations with courses matched on total pulse dose. The accelerated and non-accelerated arms of this study were equivalent in response rates and time to response. These data suggest that accelerating rTMS delivery did not reduce clinical efficacy or increase adverse events. The aTBS group found stimulation more uncomfortable which is to be expected as the stimulation trains were longer and inter-train intervals shorter in the accelerated group. The authors anecdotally reported that the patients preferred the accelerated treatment schedule due to reduced number of visits to the clinic (35). Similar research comparing accelerated and conventional TBS with all stimulation parameters constant except stimulation schedule is needed.

The most accelerated aTBS protocol for MDD to date has been the Stanford Accelerated Intelligent Neuromodulation Therapy (SAINT) protocol (21,22,28). This protocol delivers the same number of pulses as the conventional iTBS 6-week protocol in just one day. This protocol is 5 days long and therefore delivers a total of 90,000 pulses of iTBS, five times the conventional iTBS dose. 10 sessions are delivered in a single-day with a pulse intensity of 90% rMT with a depth correction and inter-session interval of 50 minutes. In addition to the accelerated delivery and high pulse-dose, the SAINT protocol comprises personalized functional connectivity MRI-guided targeting. The region of the DLPFC most anticorrelated to the subgenual anterior cingulate cortex (sgACC) is identified for each participant and targeted using neuronavigation software. The open-label remission rate after the 5-day SAINT protocol was 90% (21) while the remission rate in the first randomized-controlled trial (RCT) was 78.6% (22). Additionally, minimal side-effects, equivalent to conventional TMS protocols, were reported (21,22). Due to the multiple stimulation parameter alterations compared to conventional iTBS, RCTs are needed to directly assess the individual contribution of each of these parameters to the clinical efficacy of SAINT.

Concerns have been raised that inducing rapid antidepressant responses with accelerated protocols may result in rapid relapses (11,36). Existing durability data for antidepressant aiTBS protocols is limited with only a small number of studies measuring clinical responses beyond the cessation of stimulation (21–23). The existing durability data suggest antidepressant responses induced by aiTBS are sustained at least 4 weeks after the cessation of stimulation (21,22). Longer-term durability data for accelerated TBS protocols is required to understand the impact that accelerated delivery has on long-term antidepressant effects. Tapering and maintenance protocols for aiTBS may also maximize antidepressant response durability (37,38). High relapse rates and the need for maintenance treatment are common in other forms of TRD treatment including ECT (39–43). The likely need for regular maintenance treatment following aiTBS, in at least a proportion of TRD patients, raises current financial and accessibility issues (44–46). Currently, aiTBS protocols are not covered by insurance providers in the United States, meaning these treatments can be very costly. To access these treatments, patients also need to live in the vicinity of a TMS clinic offering these services or have the financial means to travel to and stay in an area that has one of these clinics, for the duration of treatment.

Clinical efficacy of aTBS for other conditions

In addition to the data for aTBS in MDD, aTBS trials have also been run in individuals with schizophrenia (n=3)(12,47,48), OCD (n=3)(17,49,50), Parkinson’s disease (n=2) (51,52), healthy adults (n=2)(53,54), postpartum depression (n=1)(55), post-stroke hemiparesis (n=1)(56), Alzheimer's disease (n=1)(57), methamphetamine dependence (n=1)(58), existential distress (n=1)(59), smoking cessation (n=1)(15), cocaine use disorder (n=1)(14) and bipolar disorder (n=5) (26–29). Four of the trials in patients with bipolar disorder and one trial in patients with schizophrenia also included patients with MDD, so were also included in the section above. See Table 2 for a full list of aTBS studies in patient populations other than MDD. The majority of these studies (64%) stimulated the left DLPFC, although 2 of these LDLPFC studies also stimulated the right DLPFC. Other stimulation sites in these aTBS studies include the motor cortex (54,56), the orbitofrontal cortex (50), right frontal pole (49), dorsomedial prefrontal cortex (47), cerebellum (48) and left supplementary motor area (17,52). The majority of aTBS protocols involved iTBS, only 3 studies delivered cTBS alone without iTBS. The data from these preliminary aTBS studies for diagnoses other than depression suggest aTBS protocols could be efficacious for other conditions, including improving working memory in healthy controls (53), but these pilot studies have small sample sizes with an average of 30 participants receiving active aTBS per study. Only a small number of aTBS studies have included assessments beyond the immediate cessation of stimulation (14,15,17), the longest follow-up duration has been 5 weeks post-stimulation. Participants with OCD (n=26) were assessed 5 weeks after an accelerated 3-week, 30 session cTBS protocol and responses were sustained 5 weeks post-stimulation (17).

A clinicaltrials.gov search using the term ‘accelerated theta burst’ (May 30th 2023) shows there are 41 currently active aTBS clinical trials. These active trials are for a range of clinical indications including borderline personality disorder, OCD, cognitive deficits in cancer patients, anxiety, bipolar disorder, MDD, suicidality, and smoking cessation.

aTBS parameters that appear to be important for clinical efficacy

Accelerated TBS protocols involve much shorter inter-session intervals than the approximate 24-hour interval seen in conventional TBS or rTMS protocols, the stimulation intensities are typically lower, many aTBS studies have delivered more pulses per session, varied the total number of sessions and used different targeting methods compared to conventional TBS or rTMS protocols (see Table 1 and Table 2). All of these stimulation parameters likely influence the clinical efficacy of aTBS in comparison to the conventional TBS/rTMS protocols in addition to the accelerated delivery. A greater number of sessions and higher total pulse-dose appear to result in superior clinical efficacy (60–62). Longer inter-session intervals in aTBS protocols may also improve clinical outcomes (21,63–66). Basic neuroscience data suggest that stimulation sessions delivered 50–90 minutes apart but not 40 minutes or less, result in cumulative synaptic plasticity changes (63–65). Studies in humans have shown that iTBS sessions delivered 15 minutes apart do not increase cortical excitability further than a single iTBS session (66,67) and iTBS sessions delivered 30 minutes apart induce greater changes in cortical excitability than sessions delivered with 0 or 10-minute inter-session intervals (54). Desired functional connectivity changes between LDLPFC and sgACC following LDLPFC stimulation are seen 27 minutes post-rTMS and strongest at 45-minutes post-rTMS but not observed 10 minutes post-rTMS (68). Collectively, these data suggest that longer inter-session intervals (>15 minutes) may be important for the efficacy of aTBS protocols.

The targeting method used to locate the stimulated brain area may also be important for clinical efficacy of aTBS protocols(21,69–72). The standard method currently used in TMS clinics is scalp measurements which can result in missing the targeted brain region completely (73,74). Magnetic resonance imaging (MRI), positron emission tomography (PET) and functional MRI (fMRI) have been used to identify personalized TMS targets based on specific brain structures, foci of aberrant metabolism or activation respectively (75). These methods were not shown to substantially improve clinical outcomes beyond scalp measurement-based targeting (75). In contrast, using fMRI to identify personalized targets based on functional connectivity with the neural network implicated in the condition being treated appears to improve clinical efficacy (70,76). However, the degree to which this targeting method improves clinical outcomes is controversial (77). RCTs directly comparing the clinical efficacy of fMRI-guided aTBS and scalp measurement-based aTBS are required.

Lower stimulation intensities may be more effective, for TBS protocols, including aTBS, than the 120% rMT intensity used for conventional FDA-approved iTBS and rTMS protocols. Studies have shown greater changes in cortical plasticity following TBS sessions delivered at intensities lower than an individual’s motor threshold, compared to TBS sessions delivered at motor threshold value or higher (66,78). TBS sessions delivered at 90% rMT have been shown to induce greater desired resting-state functional connectivity changes than TBS sessions delivered at 120% rMT (79).

Randomized controlled trials are needed to systematically test the direct contribution of each individual stimulation parameter on the efficacy of aTBS protocols.

Further optimization of aTBS protocols

It may be that personalization of stimulation parameters is optimal rather than any specific stimulation parameter values for all patients. For example, it is possible that stimulating at an individual’s peak theta oscillation frequency may be optimal (80). Patients with different depressive symptom clusters may respond best to different stimulation locations (81–84). Similarly, different individuals appear to exhibit different response trajectories to TBS (22,85,86) and therefore, it is likely that individualizing the number of stimulation sessions will improve clinical outcomes as some individuals require a greater number of stimulation sessions to elicit a response than others.

Another factor that could enhance the clinical efficacy of aTBS is the identification of optimal concurrent therapies. Research to date suggests that certain medications such as N-methyl D-aspartate receptor agonists (87–89) or stimulants (90,91) could enhance TBS responses. Other medications such as benzodiazepines may hinder TBS responses (90,92,93). However, the data are inconsistent (94,95). Other concurrent therapies such as cognitive behavioral therapy (CBT) could also maximize TBS-induced effects (96–98). Further research into the mechanisms underlying TBS-induced responses is needed in order to optimize concurrent therapies for patients receiving aTBS.

Currently, there are no FDA-approved maintenance protocols for any TMS treatments and maintenance protocols published to date have been extremely variable (37,38,99). There is high inter-individual variability in the durability of TMS-induced responses and a lack of long-term follow-up data from aTBS trials and therefore, it is currently unknown when and how often to deliver maintenance treatment following an acute course of aTBS (22,100–103). Personalized treatment plans including tapering and maintenance TBS sessions will likely improve long-term outcomes for aTBS protocols (38,99). The identification of reliable biomarkers that are indicative of response durability or relapse would inform the development of personalized maintenance protocols.

Conclusion

The development of aTBS protocols presents the opportunity to deliver more stimulation sessions to patients in shorter periods of time, resulting in faster and potentially more clinically effective treatment. The majority of sTBS studies to date have been in MDD patient populations, although, preliminary data suggest aTBS protocols could be effective for other diagnoses including OCD (49) and Alzheimer’s disease (57). The existing aTBS data suggest that accelerating the delivery of TBS sessions does not increase the incidence or severity of adverse events. Protocols with a higher number of TBS sessions and overall higher pulse-dose appear to have higher clinical efficacy. Longer inter-session intervals and neuronavigated targeting may also be important to maximize clinical efficacy. RCTs are needed to directly compare the clinical efficacy of aTBS and non-accelerated TBS when all other stimulation parameters are kept constant. Long-term follow-up data is also required to determine the impact of accelerating TBS delivery on response durability. Future increased personalization of stimulation parameters, identification of optimal concurrent therapies and the development of maintenance protocols may further improve the clinical efficacy of aTBS protocols.