(223 days)
The Magnus Neuromodulation System with SAINT Technology is indicated for the treatment of Maior Depressive Disorder (MDD) in adult patients who have failed to achieve satisfactory improvement from prior antidenressant medication in the current episode.
The Magnus Neuromodulation System (MNS) with SAINT Technology is a non-invasive repetitive transcranial magnetic stimulation (rTMS) system that delivers individualized and navigationally directed repetitive magnetic pulses to the left dorsolateral prefrontal cortex (L-DLPFC) to treat Major Depressive Disorder (MDD) in adult patients who have failed to achieve satisfactory improvement from prior antidepressant medication in the current episode.
The MNS with SAINT Technology is available for prescription use only and is intended for use by trained medical professionals in either an inpatient or outpatient setting.
The Magnus Neuromodulation System consists of hardware devices (stimulator with treatment coil and neuronavigation system) intended to deliver SAINT Technology; that is, rTMS (as intermittent theta burst stimulation (iTBS)) to a target area within the L-DLPFC along with proprietary software informed by structural and functional MRI and designed to identify the individualized target within the L-DLPFC for stimulation. Also included in the system are a coil and monitor for motor threshold determination, which is used to inform patient-specific stimulation settings. SAINT Technology is the combination of using the specific individualized target for treatment along with a proprietary accelerated treatment protocol that condenses treatment to five days. The Magnus Neuromodulation System is designed to support successful delivery of SAINT Technology.
Here's a breakdown of the acceptance criteria and the study details for the Magnus Neuromodulation System (MNS) with SAINT Technology, based on the provided FDA 510(k) summary:
The document does not explicitly state "acceptance criteria" in a quantitative table for this device as it would for a diagnostic AI. Instead, the submission argues for substantial equivalence to a predicate device (Nexstim NBT System 2) by demonstrating comparable safety and effectiveness, meaning the "acceptance criteria" were met by showing performance akin to or exceeding the predicate, without introducing new safety concerns. The performance is primarily evaluated by achieving a high rate of response and remission in Major Depressive Disorder (MDD).
1. Table of Acceptance Criteria (Implied) and Reported Device Performance
| Parameter | Implied Acceptance Criteria (via Predicate Equivalence) | Reported Device Performance (SAINT Technology) |
|---|---|---|
| Effectiveness (Response Rate) | Comparable to predicate (THREE-D iTBS: 49% HDRS reduction ≥50%) | Study #1 (Open-label): 83.3% (MADRS reduction ≥50%) |
| Study #2 (Open-label): 90.5% (MADRS reduction ≥50%), 86.4% (HDRS-17 reduction ≥50%) | ||
| Study #3 (Active arm, RCT): 85.7% (MADRS reduction ≥50%) | ||
| Study #4 (Open-label): 92.8% (MADRS reduction ≥50%) | ||
| Effectiveness (Remission Rate) | Comparable to predicate (THREE-D iTBS: 32% HDRS-17 ≤7) | Study #1 (Open-label): 83.3% (MADRS ≤10) |
| Study #2 (Open-label): 90.5% (MADRS ≤10), 77.3% (HDRS-17 ≤7) | ||
| Study #3 (Active arm, RCT): 78.6% (MADRS ≤10) | ||
| Study #4 (Open-label): 78.6% (MADRS ≤10) | ||
| Safety | No new or increased safety concerns compared to predicate | Zero serious adverse events (SAEs) reported across all four SAINT studies. |
| Non-serious adverse events are similar in type and incidence to the predicate device (THREE-D iTBS), with the exception of increased fatigue (likely due to extended daily treatment time for SAINT) and potentially some back/neck pain due to time spent in clinic. | ||
| Target Identification | Aid in localization of appropriate target areas within the L-DLPFC for stimulation | Proprietary software utilizing structural and functional MRI data to locate the target area within the L-DLPFC for depression therapy. Provides more individual specificity than historically used external anatomical landmarks. |
| Stimulation Protocol | Delivery of iTBS, comparable efficacy to predicate | Optimized and compressed iTBS treatment schedule over 5 days (90,000 pulses total, 10 sessions/day) compared to predicate (18,000 pulses total, 1 session/day over 6 weeks). Magnetic field intensity at 90% Motor Threshold for SAINT vs. 120% Motor Threshold for predicate, which "may be safer and more focal, and may be more effective." |
Study Proving Acceptance Criteria (Clinical Studies of SAINT Technology)
The device's performance is demonstrated through four clinical studies using SAINT Technology, which are summarized and compared to the predicate device in the provided document.
2. Sample Sizes Used for the Test Set and Data Provenance
The "test set" in this context refers to the participant groups in the clinical trials that evaluated the SAINT Technology.
- Study #1: 6 participants.
- Study #2: 21 participants (22 recruited, 1 ineligible, 1 withdrew on Day 1).
- Study #3 (Randomized Controlled Trial): 29 participants (14 active, 15 sham).
- Study #4 (Open-label Pilot): 14 participants.
Total Participants across all SAINT studies: 70 patients.
Data Provenance: The document states that these studies were "conducted in close geographical proximity." While a specific country is not explicitly named, the context of FDA submission and US-based company suggests these were likely US-based studies. All studies appear to be prospective clinical trials, tracking patients through treatment and follow-up. The note explicitly states: "As a result of studies being performed at a single site, generalizability to the broader United States population has not been evaluated."
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
The concept of "ground truth" for a diagnostic AI (e.g., expert reads of images) doesn't directly apply here in the same way, as this device is a neuromodulation therapy for MDD, not a diagnostic tool requiring expert interpretation for diagnosis. The "truth" for its effectiveness is based on patient-reported outcomes and clinician-rated scales (MADRS, HDRS), which are standard, validated instruments in psychiatry. The administration and interpretation of these scales would typically be performed by trained clinical staff and psychiatrists involved in the studies, who serve as the evaluators of patient response to treatment. Specific numbers or qualifications of individual "experts" for this "ground truth" (i.e., MADRS/HDRS scoring) are not provided, as it's an inherent part of clinical trial methodology.
4. Adjudication Method for the Test Set
Adjudication methods (like 2+1, 3+1 for discrepancies) are typically used for establishing ground truth in diagnostic studies where multiple readers might interpret data. Since this is a treatment device whose efficacy is measured by standardized psychiatric scales, such an "adjudication method" for the test set is not explicitly described in the context of expert consensus on a diagnosis or finding. The clinicians administering rating scales would be trained and standardized to minimize inter-rater variability as part of good clinical practice. In Study #3, it was a randomized, blinded trial where response and remission were assessed by clinicians blind to treatment allocation.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done in the typical sense this term is used for AI-assisted diagnostic devices (i.e., comparing multiple human readers' diagnostic accuracy with and without AI assistance).
The studies described evaluate the effectiveness of the device (SAINT Technology) as a treatment for MDD, not as an AI-assistance tool for human readers in diagnosis. While the technology involves proprietary software for target identification, the clinical trials focus on the overall clinical outcome of the therapy itself, rather than how the AI component improves human diagnostic performance.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)
The device, the Magnus Neuromodulation System (MNS) with SAINT Technology, is a treatment system that includes proprietary software for individualized target identification. It's not a standalone diagnostic algorithm. The "SAINT Technology" itself is described as "the combination of using the specific individualized target for treatment along with a proprietary accelerated treatment protocol." Therefore, it inherently involves human medical professionals for its application ("intended for use by trained medical professionals"). The performance data presented are for the entire system in a clinical treatment context, not for the algorithm in isolation.
7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)
The "ground truth" for the effectiveness of the SAINT Technology is established through clinical outcomes data using validated psychiatric rating scales and predefined criteria:
- Montgomery-Asberg Depression Rating Scale (MADRS):
- Response: ≥50% reduction in MADRS score.
- Remission: MADRS score ≤10.
- Hamilton Depression Rating Scale (HDRS-17) / (HDRS-6):
- Response: ≥50% reduction in HDRS score.
- Remission: HDRS-17 score ≤7 or HDRS-6 score ≤4.
Safety is assessed by monitoring and reporting Adverse Events (AEs), including Serious Adverse Events (SAEs).
8. The Sample Size for the Training Set
The document does not specify a separate "training set" for the clinical performance evaluation of the device in the way a machine learning model would have one. The clinical studies (1-4) are the evaluation (test) sets for the device's clinical efficacy and safety.
However, the "proprietary software informed by structural and functional MRI" that "identify the individualized target within the L-DLPFC for stimulation" would have, by its nature, been developed and trained using various datasets. The size and nature of such internal development/training datasets for the software component are not provided in this summary.
9. How the Ground Truth for the Training Set Was Established
As mentioned above, information regarding a specific "training set" for the software component is not detailed in this 510(k) summary. For software that generates individualized targets, the "ground truth" for its development would typically be established based on:
- Neuroanatomical models and expert neuroimaging knowledge: Defining correct anatomical and functional regions in the brain.
- Clinical correlation: Data linking specific MRI features (structural and functional connectivity patterns) to treatment response in previous studies (either published literature or internal pilot data).
This summary focuses on the clinical validation of the entire system as a treatment, assuming the target identification software was developed using robust, scientifically sound methods prior to these clinical trials.
§ 882.5805 Repetitive transcranial magnetic stimulation system.
(a)
Identification. A repetitive transcranial magnetic stimulation system is an external device that delivers transcranial repetitive pulsed magnetic fields of sufficient magnitude to induce neural action potentials in the prefrontal cortex to treat the symptoms of major depressive disorder without inducing seizure in patients who have failed at least one antidepressant medication and are currently not on any antidepressant therapy.(b)
Classification. Class II (special controls). The special control is FDA's “Class II Special Controls Guidance Document: Repetitive Transcranial Magnetic Stimulation System.” See § 882.1(e) for the availability of this guidance document.