FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The IlluminOss Photodynamic Bone Stabilization System (PBSS) is indicated for skeletally mature patients in the treatment of impending and actual pathological fractures of the humerus, radius, and ulna, from metastatic bone disease.

Device Description

The IlluminOss Photodynamic Bone Stabilization System (PBSS) is intended to be used in the fixation and stabilization of actual and impending pathological fractures of the humerus, radius, and ulna through a minimally invasive procedure. The system uses a catheter to deploy an inflatable, noncompliant, thin wall PET balloon into the medullary canal of the bone across the fracture site. The balloon is infused using a standard 20cc syringe with a photodynamic (light cured) monomer that causes the balloon to slowly expand and fill the intramedullary canal of the fractured bone. Activation of the light system allows for visible spectrum light to be delivered through a radially emitting light pipe that is temporarily positioned within a central lumen of the catheter that runs the length of the balloon. The liquid monomer within the balloon is exposed to light along the entire length of the balloon during the curing process. Curing (and hardening) occurs only when the photo initiator within the monomer is exposed to a specific frequency of light causing rapid polymerization of the monomer resulting in a solid intramedullary (IM) rod. The time to cure the IM rod depends on the size of the balloon used to stabilize the fracture. A Timer Kev, included within each balloon catheter kit, determines the time the light source is activated during the curing process to ensure the appropriate cure time is used for each balloon size.

AI/ML Overview

Here's an overview of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

The provided document extensively details various tests and their acceptance criteria for the IlluminOss PBSS. Many of these relate to direct physical and chemical properties of the device components. The clinical study also had specific efficacy and safety endpoints.

Due to the volume of detailed bench testing, the table below will focus on the clinical efficacy and safety endpoints, as well as a selection of representative non-clinical acceptance criteria.

Category	Specific Test / Criterion	Acceptance Criteria	Reported Device Performance
Clinical Efficacy	Primary Efficacy Endpoints (Day 90 vs. Baseline):
	- VAS Pain Score Reduction (non-inferiority)	Mean improvement in VAS over 90 days > 80% of historical reference. (Historical Ref. improvement of >53.8)	The study showed an improvement of about 79% of that seen with the control, failing to statistically meet this co-primary efficacy endpoint. However, a significant clinical improvement in pain relief over baseline was noted (average reduction of 53 points from baseline to Day 90).
	- MSTS Functional Score Improvement (non-inferiority)	Mean improvement in MSTS over 90 days > 80% of historical reference. (Historical Ref. improvement of >23.7)	The study successfully met this co-primary efficacy endpoint, with a 40-point increase in MSTS from baseline to Day 90. Patients with the device appear comparable to controls in overall clinical function.
Clinical Safety	Primary Safety Endpoint (Day 90):	Meeting all of the following: no serious device-related adverse events; no additional surgical interventions (revision, removal, supplemental fixations); no device fracture, migrations, mal-alignment, or loss of reduction or fixation as evidenced by radiographic review.	The clinical trial results validated the ability of the PBSS to safely stabilize a fracture. Device/Procedure related adverse events were low (n=32), with only five subjects requiring a second surgery related to the device. No bone infections were observed, and only one wound site infection occurred. The study successfully met its primary safety endpoint.
Biocompatibility	Cytotoxicity	Non-cytotoxic	Non-cytotoxic
	Sensitization	Non-sensitizing	Non-sensitizing
Mechanical	Balloon Catheter Burst Pressure	Mean burst pressure must meet internal acceptance criterion. No individual sample below acceptance criterion.	The mean burst pressure of every OD balloon catheter assembly met the internal acceptance criterion. No individual test sample exhibited burst pressure results below the acceptance criterion.
	IM Rod Static Four-Point Bend Strength (compared to legally marketed device)	Comparable to a legally marketed device with similar indications.	Compared very favorably to an existing, marketed device. IlluminOss device possesses adequate mechanical properties.
	Cure Validation - Shore D Hardness (post-cure)	All quadrants in all sections must meet the required Shore D Hardness specification.	All quadrants in all sections met the required Shore D Hardness specification. Full cure achieved in less time than coded timer keys.
Electrical/EMC	EMC Testing (e.g., Radiated Emissions, Conducted Emissions, ESD Immunity, Radiated RF Immunity, Electrical Fast Transient/Burst Immunity, Surge Immunity, Conducted Disturbances Immunity, Power Magnetic Field Immunity) for Photodynamic Light Source	Specific limits (e.g., Class B Below Limit) for emissions tests. For immunity tests: The light source shall continue to operate as intended without operator intervention. For voltage dips/short interruptions: The output of light may be momentarily interrupted. However, the light output must resume, without operator interventions.	Complied for all emission tests. Passed all immunity tests, with no degradation of performance observed during or after application of test voltages/fields. Passed voltage dip/interruption test, with light output resuming without operator intervention.
Sterility	Sterility Assurance Level (SAL) for device components	SAL of 10^-6	Achieved a SAL of 10^-6 (via EO sterilization).
	Aseptic Media Fill Process Validation	When filling fewer than 5,000 units, no contaminated units. A minimum of 3 successive annual Media Fill Process Simulations illustrating uncontaminated units at batch sizes consistent with manufacturing runs.	The acceptance criteria were met, validating the Media Fill process.
Shelf Life	Package integrity, system functionality, and sterility over time	Confirmed a three (3) year shelf life.	Testing confirmed a three (3) year shelf life was supported based on real time and accelerated aged product. Implant and disposable delivery kit, including monomer, are labeled with a three-year shelf life.
Thermal Safety	Temperature During Cure	Acceptable temperature profile during the curing process that minimizes the risk of tissue injury.	Animal testing demonstrated no short or long-term adverse effects due to exothermic reaction. Clinical results from 81 IDE study patients found no adverse events attributed to exothermic reaction. Test results exhibit an acceptable temperature profile.

2. Sample Size Used for the Test Set and Data Provenance

Test Set Sample Size (Clinical Study): 81 subjects.
Data Provenance:
- Country of Origin: The IDE clinical study was a multi-center study conducted at 13 sites, presumably in the United States, given it was an Investigational Device Exemption (IDE) study within the US regulatory framework.
- Retrospective or Prospective: The IDE clinical study was a prospective study. It also compared its results to historically controlled (literature-based) data for its non-inferiority assessment. An earlier EU Registry Study (135 bones in 132 subjects) was also mentioned, which collected data prospectively (standard of care demographic and fracture-related data for adverse device effects), but not as a formal clinical trial for this De Novo submission.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

For the clinical study, ground truth was based on patient-reported outcomes (VAS pain, MSTS functional score, etc.), radiographic evaluations, and adverse event reporting.

Number of Experts: Not explicitly stated as a specific number of independent experts providing a "ground truth" adjudication for each case. Outcomes were measured by study personnel and clinical investigators at the participating sites.
Qualifications of Experts: Clinical Investigators at the 13 multi-center sites would be qualified physicians and surgeons involved in the treatment of patients with pathological fractures and musculoskeletal oncology. The radiographic evaluations for primary safety endpoints (device fracture, migrations, mal-alignment, or loss of reduction or fixation) would have been performed by qualified radiologists or orthopedic surgeons.

4. Adjudication Method for the Test Set

For the clinical study, the adjudication method involved:

Pain and Function: Patient-reported outcome measures (PROMs) like VAS and MSTS scores were collected directly from subjects.
Safety Endpoints: Assessment of major device-related adverse events, additional surgical interventions, and radiographic evaluations. The "radiographic evaluations" likely involved clinical investigators reviewing images, but a multi-reader adjudication process (e.g., 2+1 or 3+1) is not explicitly described for these. Adverse events would typically be classified and confirmed by the study investigators and potentially reviewed by an independent Clinical Events Committee (CEC), but this is not detailed in the provided text.

For non-clinical bench testing, the "ground truth" was established by comparing test results against predefined engineering and performance standards/criteria (e.g., ISO, ASTM, internal specifications).

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, If So, What was the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done.
This device is an implantable medical device for fracture stabilization, not an AI diagnostic or assistive tool for human readers. Therefore, the concept of "human readers improve with AI vs. without AI assistance" does not apply to this submission.

6. If a Standalone (i.e., Algorithm Only Without Human-in-the-loop Performance) Was Done

No, a standalone (algorithm only) performance study was not done.
Again, this is a physical medical device, not a software algorithm or AI. Its performance is demonstrated through bench testing (mechanical properties, curing, etc.) and clinical outcomes in patients.

7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)

For the clinical study of the IlluminOss PBSS, the ground truth was primarily based on:

Patient-Reported Outcomes (PROs): VAS pain scores, MSTS functional scores, QLQ-BM22 pain and functional interference, pain at palpation, analgesic use, and ability to perform Activities of Daily Living (ADLs).
Clinical Outcomes/Safety Data: Incidence of adverse events, need for additional surgical interventions, and radiographic evidence of device fracture, migration, mal-alignment, or loss of reduction or fixation.
Comparison to Historical Controls: The efficacy endpoints used historical literature as a "reference" for non-inferiority comparisons, which can be interpreted as a form of expert consensus or aggregated outcomes data from prior studies.

For non-clinical bench studies, the ground truth was based on:

Established Test Standards: Adherence to ISO, ASTM, and other recognized engineering standards and internal specifications (e.g., Shore D Hardness, burst pressure, tensile strength limits, EMC compliance).
Animal Studies: Observed biological reactions (e.g., tissue necrosis due to exothermic reaction) and successful implantation/removal in cadaveric models.

8. The Sample Size for the Training Set

The concept of a "training set" primarily applies to machine learning algorithms.

For the clinical study, a training set in the AI sense is not applicable as it was a traditional clinical trial.
For bench testing, there isn't a "training set" in the machine learning sense. Instead, product development and iterative design likely involved numerous prototypes and tests to optimize the device's performance before formal validation testing. The "Timer Key" development for cure times could be seen as an optimization based on prior experimentation for different balloon sizes.

9. How the Ground Truth for the Training Set Was Established

As noted above, a "training set" as understood in AI/ML is not directly applicable here. The "ground truth" for the development and optimization of the device (leading to its final design that then underwent formal testing) would have been established through a combination of:

Engineering Principles and Materials Science: Basic understanding of polymer chemistry, mechanics, and biocompatibility.
Pre-clinical Research and Development: Iterative design, prototyping, and testing in laboratory settings to meet desired performance characteristics, often guided by
standards and clinical needs.
Biocompatibility Testing: According to ISO 10993 series for an implanted device.
Previous Experience/Literature: The very existence of similar legally marketed devices and the EU Registry study provided context and informed the design and expected performance, forming a basis for "what good looks like."

Summary

Regulation Number and Section

§ 888.3023 In vivo cured intramedullary fixation rod.

(a)
Identification. An in vivo cured intramedullary fixation rod is a prescription implanted device consisting of a balloon that is inserted into the medullary canal of long bones for the fixation of fractures. The balloon is infused with, and completely encapsulates, a liquid monomer that is exposed to a curing agent which polymerizes the monomer within the balloon creating a hardened rigid structure.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Non-clinical performance testing must demonstrate that the device performs as intended under anticipated conditions of use. The following performance characteristics must be tested:
(i) Mechanical testing must be conducted on the final device to assess burst, abrasion, bending, and torsion in static and dynamic conditions.
(ii) Mechanical testing must demonstrate the integrity of the balloon including testing for leaks, ruptures, and release of cured/uncured material.
(iii) Performance testing must demonstrate that the device can be inserted and removed.
(iv) Performance testing must demonstrate the ability, in the event of a leak, to remove the uncured material from its in vivo location.
(v) Performance testing must demonstrate the reliability and accuracy of the curing method used.
(vi) Thermal safety testing must be conducted to evaluate the temperature rise during curing.
(2) Electrical safety, electromagnetic compatibility (EMC) testing, and electromagnetic interference (EMI) testing must be conducted for all electrical components.
(3) All patient-contacting components must be demonstrated to be biocompatible.
(4) Performance data must demonstrate the sterility and pyrogenicity of patient contacting components of the device that are provided sterile.
(5) Performance data must validate the reprocessing instructions for any reusable components or instruments.
(6) Performance data must support the shelf life of the system by demonstrating continued sterility, package integrity, and system functionality over the established shelf life.
(7) Technological characterization of the device must include materials, curing agents, and a description of the operating principles of the device, including the delivery system and devices which initiate the curing process.
(8) Labeling must include the following:
(i) A detailed summary of the device technical parameters.
(ii) Information describing all materials of the device.
(iii) Information describing how to perform the procedure and use the device, including the delivery system and devices which initiate the curing process, as well as how to remove the device and any uncured materials.
(iv) A shelf life.
(v) Validated methods and instructions for reprocessing any reusable components or instruments.