(234 days)
The PoNS device is indicated for use as a short term treatment of gait deficit due to mild to moderate symptoms from multiple sclerosis and is to be used as an adjunct to a supervised therapeutic exercise program in patients 22 years of age and over by prescription only.
The PoNS™ device is a non-implantable tongue stimulator that delivers electrical stimulation to the trigeminal and facial nerves for use as a short-term treatment of gait deficit due to mild to moderate symptoms from multiple sclerosis and is to be used as an adjunct to a supervised therapeutic exercise program. The device is for use in patients over 22 years of age by prescription only.
The PoNS™ mouthpiece is held lightly in place by the lips and teeth around the neck of the tab that goes into the mouth and rests on the anterior, superior part of the tongue. The control unit is worn around the neck during session usage. It has 5 user-controlled buttons: power, information, intensity up, intensity down, and start/stop/pause. The control unit also records usage data including session duration and general activities performed which are captured by an accelerometer. During a patient's visit with a therapist is able to connect the control unit to a computer and, using software developed specifically for the PONSTM, view the usage data. The purpose of the usage data is to give the therapist information on how to improve a patient's execution of therapy by identifying potential areas of missed or shortened sessions. The PoNSTM System is comprised of a Controller, a Mouthpiece Retainer Case, a Charger, a Product Carry Case and the PoNSTM Software. Accessory components, also supplied, include a cable for connection to a computer running the PoNSTM software.
Controller: The Controller includes the primary user interface and drive electronics for providing the electronic stimulation waveform. Through the user interface, the intensity of the stimulation waveform can be adjusted, the stimulation can be started and stopped, and the Controller can be powered on and off. The Controller user interface consists of a visual display, audio feedback and vibration feedback.
Mouthpiece: The Mouthpiece houses the electrode array through which the stimulation waveform is applied to the patient's tongue. Electronics and software within the Mouthpiece control the timing of the stimulation waveform. The Controller sends commands to the Mouthpiece and receives status messages from the Mouthpiece via a cable that forms an integral part of the Mouthpiece assembly.
Mouthpiece Retainer Case: The Mouthpiece Retainer Case is used to store the Mouthpiece between uses. The cable of the Mouthpiece wraps around the base of the Mouthpiece Retainer Case. There is a pocket in the Carry Case sized to house the Mouthpiece Retainer Case during storage.
Charger: The Charger consists of a mains power adaptor that connects to the Controller via a standard USB Type-C connector and provides a low voltage power supply to recharge the Controller battery. Stimulation is disabled while the Charger is connected to the Controller.
Based on the provided text, the Portable Neuromodulation Stimulator (PoNS) is not an AI/ML powered device, therefore the typical acceptance criteria and study designs (e.g., MRMC studies, standalone AI performance, expert ground truth establishment for AI training/test sets) for such devices are not directly applicable.
The document describes the regulatory approval of a medical device based on non-clinical (bench testing, biocompatibility, electrical safety, software validation) and clinical performance. The "acceptance criteria" discussed are primarily related to safety, functionality, and clinical effectiveness demonstrated through traditional clinical trials.
However, I can extract the relevant information from the provided text to fulfill as much of your request as possible, reinterpreting "acceptance criteria" in the context of this device's approval for clinical use and "study that proves the device meets the acceptance criteria" as the evidence presented for its De Novo classification.
Here's the breakdown:
Acceptance Criteria and Device Performance (Interpreted for a Non-AI Device)
The "acceptance criteria" for the PoNS device, as inferred from the regulatory approval process described, are primarily focused on demonstrating safety, biocompatibility, electrical and mechanical performance, and clinical effectiveness for its indicated use.
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria Category | Specific Criteria/Requirements (Inferred) | Reported Device Performance / Evidence Presented |
|---|---|---|
| Biocompatibility | All patient-contacting components are biocompatible. | "The PoNSTM mouthpiece and components is categorized as a surface device in prolonged (24 hours to 30 days) contact of the mucosal membrane (i.e. mouth, tongue lips). Assessment of the device included the following tests which met the prespecified acceptance criteria: Cytotoxicity, Sensitization Test, Intracutaneous Reactivity, Oral Mucosa Irritation, Acute Systemic Toxicity, Material-mediated pyrogenicity, Extractables & Leachables Study, Chemical Characterization of medical device materials, Genotoxicity." |
| Electrical, Mechanical, Thermal Safety | Electromagnetic compatibility (EMC), battery safety, and electrical, mechanical, and thermal safety of the device must be demonstrated. | EMC/Electrical Safety: Tested according to IEC 60601-1:2005, IEC 60601-1-2:2014, IEC 60601-1-11:2015, IEC 60601-2-10:2012. "Results demonstrated that the device is compliant to this standard." |
| Battery Safety: "The battery complies with IEC 62133-2 (Standard for lithium batteries), and IEC 60601-1 (Basic Safety)." | ||
| Electrical Stimulation Output: "Testing was performed to characterize the stimulation output waveform, the functionality of the PoNSTM as a system, and the requirements of the output stimulation parameters. Results demonstrated that the system meets specifications." | ||
| Electrode Bench Testing: "Testing was performed to assess the mechanical measurements, the design of the electrodes... and the electrical characteristics... Results demonstrated that the electrodes passed all testing." | ||
| Software Validation & Hazard Analysis | Software verification, validation, and hazard analysis must be performed, with documentation including cybersecurity assessment. | "Software of this device is considered to have a 'Moderate' level of concern." "The submission contained all the elements of software documentation corresponding to a 'Moderate' level of concern... Adequate documentation... provides the foundation that the software will operate in a manner as described in the specifications. A hazard analysis was performed to characterize software risks... The submission included verification and validation (V&V) testing to address the potential hazards with satisfactory result." |
| Clinical Effectiveness | Demonstrated clinically meaningful improvement in gait deficit due to mild to moderate MS symptoms when used as an adjunct to a supervised therapeutic exercise program. (Primary outcome: Dynamic Gait Index (DGI) score) | Tyler et al. (2014) Study: Randomized, double-blind controlled trial. Active PoNS group achieved an average improvement in DGI score of b at b weeks, while the control group improved [0] points. "Compared to baseline, the active group achieved a statistically significant and clinically significant (4-point change in DGI) by the end of the study, while the control group did not." |
| Leonard et al. (2017) Study: Patient-blinded, randomized, controlled pilot study. Found "a nonsignificant trend towards the Active group" for DGI. Showed statistically significant improvement in SOT scores for the Active group. | ||
| Real-World Data (RWD): Retrospective analysis of Canadian MS patients. Mean improvement in FGA score was [0] points at Week [0] (n=36). Sponsor attempted pooling with Tyler/Leonard studies (adjusted DGI), but FDA notes this conversion is "not validated according to any published literature." | ||
| Safety Profile (Clinical) | Acceptable safety profile with temporary adverse events and no serious device-related adverse events. | "The device exhibited an acceptable safety profile in the clinical studies which were conducted, and any adverse events that occurred were temporary and had complete resolution. No device-related serious adverse events were observed, such as electrical injury, burns, or permanent injury to tongue nerve or muscle function." |
| Labeling | Comprehensive labeling including technical parameters, instructions for use, cleaning, storage, charging, and disposal instructions. Also, contraindications, warnings, precautions, and risk mitigation. | "The labeling (User Instructions) meets the requirements of 21 CFR Part 801.109 for prescription devices." Specific contraindications, warnings, and precautions are listed in detail in the provided text. Instructions for use, cleaning, storage, charging, and disposal are also included. |
Study Details (Clinical Evidence for Effectiveness and Safety)
The primary clinical evidence for the PoNS device's effectiveness and safety came from a combination of controlled clinical trials and real-world data.
2. Sample Sizes and Data Provenance
- Tyler et al. (2014) Study:
- Sample Size: "I subjects" (Note: the exact number 'I' is not specified or blanked out, but the context suggests a controlled trial size). The table provided later had an "N" column, but the values were blanked out (b)(4).
- Data Provenance: Not explicitly stated, but the study design "Randomized, double blind controlled trial" suggests a prospective clinical trial. Country of origin not specified.
- Leonard et al. (2017) Study:
- Sample Size: "n=7" for PoNS+cognitive/physical rehabilitation group, "n=7" for sham device+cognitive/physical rehabilitation group. (Total n=14).
- Data Provenance: Not explicitly stated, but described as a "Patient blinded, randomized, controlled trial," indicating a prospective clinical trial. Country of origin not specified.
- Real-World Data (RWD):
- Sample Size: "n=36" for the primary outcome analysis for FGA improvement. "Do of the of real world patients had baseline Expanded disability status scale (EDSS) scores" (numbers blanked out).
- Data Provenance: "retrospective analysis of real-world data ('RWD') collected with the PoNS™ in MS patients in clinical rehabilitation settings in Canada." (Retrospective, Canada).
3. Number of Experts and Qualifications for Ground Truth
This device's "ground truth" for clinical effectiveness is based on clinician-scored indices and objective measures of gait and balance in patients with Multiple Sclerosis. The document does not specify a number of individual experts or their qualifications establishing "ground truth," but rather references standard clinical assessment methods.
- Dynamic Gait Index (DGI): "clinician scored index of @gait tasks." (The specific number of tasks is blanked out). This implies that a qualified clinician administered and scored the DGI.
- Functional Gait Assessment (FGA): Also a clinical assessment of gait function.
- Sensory Organization Tests (SOT): Objective balance tests.
4. Adjudication Method for the Test Set
Not applicable in the traditional sense of AI model ground truth adjudication. Clinical trial outcomes are typically managed per study protocols, which would define how assessments (like DGI, FGA) are performed and scored by clinicians. There's no mention of an adjudication process by multiple independent experts for reconciling conflicting scores for the primary outcome measures.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No. An MRMC study is a specific type of study design typically used for evaluating the effectiveness of diagnostic tools (often imaging-based AI) by comparing performance with and without AI assistance across multiple readers and cases. The PoNS device is a therapeutic device, not a diagnostic one, and is not AI-powered, so an MRMC study is not relevant here. The clinical studies compared the device's therapeutic effect (PoNS + exercise vs. control/sham + exercise), not human reader performance with or without AI.
6. Standalone Performance
No. This concept is typically applied to AI algorithms to evaluate their performance without human interaction or a human-in-the-loop. Since PoNS is a physical medical device for treatment, its "performance" is inherently tied to its effect on the patient after being used, often in conjunction with a supervised exercise program. Its effectiveness is evaluated in a human subject (patient) in clinical trials.
7. Type of Ground Truth Used
The "ground truth" for evaluating the clinical effectiveness of PoNS was:
- Clinician-scored indices: Dynamic Gait Index (DGI) and Functional Gait Assessment (FGA). These are standardized clinical assessment tools administered and scored by healthcare professionals.
- Objective measures: Sensory Organization Tests (SOT) provided quantitative data on balance.
- Patient Outcomes: Improvement in gait deficit as measured by these tools.
There is no mention of pathology or other biological outcomes as primary ground truth, as the indication is for gait improvement in MS symptoms.
8. Sample Size for the Training Set
Not applicable. The PoNS device is not an AI/ML algorithm that requires a "training set" in the computational sense. Its "training" or development involved engineering, bench testing, and traditional clinical trials to validate its safety and effectiveness.
9. How the Ground Truth for the Training Set Was Established
Not applicable, as there is no AI "training set" for this device. Its design was based on principles of neuromodulation, and its efficacy established through clinical studies using the ground truth methods described in point 7.
N/A