(380 days)
The AeroForm Tissue Expander System is used for soft tissue expansion in breast reconstruction following mastectomy, for the treatment of underdeveloped breasts, and for the treatment of soft tissue deformities in the breast.
The AeroForm Expander is intended for temporary subcutaneous or submuscular implantation and is not intended for use beyond six months.
The AeroForm® Tissue Expander System is comprised of the AeroForm Tissue Expander, the Dosage Controller, and the Physician Master Key. The AeroForm Tissue Expander is a sterile implant with an outer textured silicone shell and a non-distensible inner bag which are anatomically shaped to allow for directional expansion in the lower, anterior pole. The AeroForm Tissue Expander contains a reservoir of compressed carbon dioxide (CO2), which is released within the AeroForm Tissue Expander by using the Dosage Controller. The outer shell has suture tabs to allow fixation of the device to the surrounding tissues to prevent rotation. A receiving antenna and electronics within the AeroForm Tissue Expander enable communication with the Dosage Controller. The AeroForm Tissue Expander has no intrinsic electrical power, batteries, or software, and can only be activated by the Dosage Controller. There are 3 different sizes of the expander, including small (400cc), medium (600cc), and large (850cc). The Dosage Controller is a small, hand held battery-powered, non-sterile device. It activates the AeroForm Tissue Expander to release the programmed amount of CO2 gas (10cc). The Dosage Controller is configured to provide coded instructions to its bonded AeroForm Tissue Expander. It has a single push button for ease of use and a bank of indicator lights (LEDs) and tones. The Physician Master Key is used only by the physician in the operating room or in the physician office. With the Physician Master Kev inserted into the Dosage Controller receptacle, the Dosage Controller can activate the Expander with no daily limit on the expansion volume.
The AeroForm® Tissue Expander System underwent extensive testing, encompassing biocompatibility, shelf-life, electromagnetic compatibility, software verification, and performance bench testing. A pivotal clinical study, XPAND trial (CPT-0003), and a Continued Access Study (CAS) provided critical insights into its performance and safety in a real-world setting.
1. Acceptance Criteria and Reported Device Performance
The acceptance criteria for several non-clinical tests and the corresponding reported device performance are summarized below:
| Test | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| Biocompatibility | ||
| Cytotoxicity (ISO MEM Elution) | The biological response of the test samples extract must be grade 2 (mild or less) | Pass |
| Sensitization (Maximization) | The test samples must show a grade of 1 or less when compared to the control | Pass |
| Irritation (ISO Intracutaneous) | The difference between the mean score of the test samples and control groups, in terms of erythema and edema, must be less than 1.0 | Pass |
| Acute Systemic Toxicity (USP Systemic Injection) | The test mice must not show a significantly greater reaction than the control mice | Pass |
| Acute Systemic Toxicity (USP Mediated Pyrogen) | The test rabbits must not show a temperature difference from the control of greater than or equal to 0.5°C | Pass |
| Sub-chronic toxicity | Evidence of irritation in the test animals will be scored and compared to the control animals | Pass |
| Chronic Toxicity | Evidence of irritation in the test animals will be scored and compared to the control animals | Pass |
| Genotoxicity (Bacterial Reverse Mutation) | The mean number of test revertants must be less than 2x the mean number of negative control revertants | Pass |
| Genotoxicity (Mouse Micronucleus) | The average % MN-RET for the test group must be less than 1.0% | Pass |
| Genotoxicity (Mouse Lymphoma) | The mutant frequency of the test sample must be 2X or less than that of the control | Pass |
| Implantation (2 week and 9 week) | The average macroscopic scores for test sample sites compared to control sites are correlated to a grade. The microscopic findings showing cellular changes will be graded according to severity (0-4) and any resulting irritant response indicated by a difference between test and control values will be graded as nonirritant, slight, moderate or severe. | Pass |
| Particulate Testing | Particle size [Redacted] (b) (4) particles per mL | |
| Particle size [Redacted] (b) (4) particles per mL | ||
| (per EN45502-1) | Pass | |
| LAL (Limulous Amoebocyte Lysate) Testing | Detected EU/device level is below (b) (4) EU/device | Pass |
| Carcinogenicity | Analysis of the biological test results, chemical characterization data and literature review | Pass by analysis and risk assessment |
| Performance - Bench Testing | ||
| Cycle Rub Test | There are no major leaks or tears in the film or the seals, after 25,000 cycles. | The Expander successfully met the no leak and tear requirements after 25,000 cycles. |
| Impact Test | There are no ruptures after the impact test. | No ruptures were reported after the impact test. |
| Impact Force Test | The device shall not fail or leak when experiencing forces less than 1700 lbs. | All acceptance criteria were met. The reservoir and valve assembly withstood an impact load of ~4X the human tolerance levels of 1700lbs without leakage. Impact loads to ~7X the human tolerance level resulted in slow leaks. |
| Leak Test | The CO2 concentration shall not increase more than 100 PPM. | The Expander successfully met the leak test requirements following functional testing. |
| Permeability Test | Small & Medium Size: Each test article shall have a daily average permeation rate of 0-4.1 mL CO2 per day. Large Size: Each test article shall have a daily average permeation rate of 0-6.0 mL CO2 per day. | Small and medium test articles met the daily average permeation rate of -10% compared to saline group. |
| Median Days to Complete Expansion | Not explicitly stated as acceptance criteria, but a key secondary effectiveness endpoint. | 21.0 days (95% CI 15.0, 24.0) |
| Median Days to Complete Reconstruction | Not explicitly stated as acceptance criteria, but a key secondary effectiveness endpoint. | 108.5 days (95% CI 99.0, 117.0) |
| Subject Satisfaction (Overall) | Not explicitly stated as acceptance criteria, but a key secondary effectiveness endpoint. | 78% (AeroForm overall), 84% (V2.5 only) |
| Physician Satisfaction (Overall) | Not explicitly stated as acceptance criteria, but a key secondary effectiveness endpoint. | 68% (AeroForm overall), 83% (V2.5 only) |
| Adverse Event Rates (Any) | Not explicitly stated as acceptance criteria, but a key safety endpoint for comparison. | 63.3% (AeroForm arm) vs 63.5% (Saline arm) |
| Device-Related AE Rates | Not explicitly stated as acceptance criteria, but a key safety endpoint for comparison. | 21.2% (AeroForm arm) vs 19.2% (Saline arm) |
| Device Malfunctions (V2.5 only) | Reduction in malfunction rates, particularly under expansion/deflation, after design modification. | Under expansion/deflation reduced from 46.7% (V2.0) to 15.4% (V2.5). |
| XPAND II CAS Treatment Success Rate | 100% (n=18 breasts evaluated) | |
| XPAND II CAS Subject Satisfaction | 89% (8/9 subjects) | |
| XPAND II CAS Physician Satisfaction | 90% (9/10 physicians) for device performance; 80% (8/10) for expansion results. |
2. Sample Size and Data Provenance
Pivotal Study (XPAND Trial - CPT-0003):
- Sample Size (Test Set - Effectiveness Analysis): 151 patients (99 AeroForm patients and 52 saline patients for effectiveness analysis). The analysis was per breast: 155 AeroForm breasts and 83 Saline breasts.
- Sample Size (Test Set - Safety Analysis): 158 subjects enrolled (106 randomized to AeroForm, 52 to Saline). One AeroForm subject was randomized but no device implanted, so 157 subjects for safety analysis.
- Data Provenance: Prospective, multi-center, randomized, controlled, open-label study conducted at 17 investigational sites in the U.S.
Continued Access Study (XPAND II CAS):
- Sample Size (Test Set): 28 subjects (49 breasts) enrolled as of June 8, 2016. Data available for 11 subjects (21 breasts) who completed the final study visit.
- Data Provenance: Prospective, multi-center, single-arm study of the V2.5 device.
Prior Investigations (Australia):
- Sample Size: 40 subjects (71 AeroForm Tissue Expanders) in the PACE 1 and PACE 2 studies. 21 subjects (34 AeroForm Expanders) in a second Australian study.
- Data Provenance: Two-phase single-center, prospective, open-label clinical study (PACE) in Perth, Australia. Second study also conducted in Australia. These were prior investigations, likely used for early feasibility and design refinement.
3. Number of Experts and Qualifications for Ground Truth
The document does not explicitly state the number of experts or their specific qualifications (e.g., radiologist with 10 years of experience) used to establish ground truth for image-based assessments. However, the ground truth for the clinical study was based on:
- Treatment Success: Defined by successful tissue expansion and exchange to a permanent breast implant, unless precluded by a non-device-related event. This assessment would have been made by the investigating physician (surgeon) at each site.
- Adverse Events: Clinical observations and diagnoses made by the investigator (physician).
- Patient-reported outcomes: Patient satisfaction and pain ratings, which are self-reported.
- Physician satisfaction: Reported by the investigating physicians.
These are clinical outcomes determined by the treating physicians and patient self-assessment, rather than independent expert consensus on specific objective findings like image interpretation.
4. Adjudication Method for the Test Set
The document does not specify a formal adjudication method (e.g., 2+1, 3+1) for the clinical outcomes in the pivotal study or the CAS. The clinical outcomes (treatment success, adverse events) were reported by the investigators at each site. Given the nature of breast reconstruction outcomes, these are typically determined by the operating surgeon and clinical staff based on direct patient assessment and surgical findings, rather than independent adjudication of ambiguous findings.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No Multi-Reader Multi-Case (MRMC) comparative effectiveness study was mentioned. The pivotal study was a direct comparison between the AeroForm Tissue Expander System and standard saline tissue expanders, both used by human clinicians. The comparison was of device performance and patient outcomes, not of human reader improvement with or without AI assistance, as the device itself is not an AI diagnostic tool.
6. Standalone (Algorithm Only) Performance
The AeroForm Tissue Expander System is a physical medical device (implant, controller, master key), not a diagnostic algorithm or AI system. Therefore, a standalone (algorithm-only) performance study is not applicable and was not conducted. The software component manages the device's function (e.g., CO2 dosing, communication) and its validation focused on software reliability, safety, and hazard analysis, as outlined in the "SOFTWARE" section.
7. Type of Ground Truth Used
- Clinical Studies (XPAND Trial & CAS): The ground truth for effectiveness was primarily based on clinical outcomes data, specifically whether tissue expansion was successful and led to exchange with a permanent breast implant. Safety ground truth was established through recorded adverse events and device malfunctions observed during the clinical course. Patient and physician satisfaction ratings were also used.
- Non-Clinical (Bench, Biocompatibility, EMC) Studies: Ground truth for these studies was established by pre-defined quantitative and qualitative acceptance criteria for specific physical, chemical, and electrical properties, as measured by standardized testing methodologies.
8. Sample Size for the Training Set
The document does not mention a separate "training set" in the context of machine learning or AI models, as the AeroForm Tissue Expander System is a physical device, not an AI/ML product.
However, if "training set" is meant to refer to early clinical data used to refine the device or inform the pivotal study design:
- PACE Studies (Australia): 40 subjects (71 AeroForm Expanders V1.0 and V2.0).
- Second Australian Study: 21 subjects (34 AeroForm Expanders).
- Sheep Implant Study: 2 male ovine subjects, 12 implants. This was a preliminary safety evaluation.
These earlier studies and animal trials served as an evolutionary process, leading to modifications (V2.0 to V2.5) that reduced malfunctions, similar to how a training set might inform model refinements.
9. How the Ground Truth for the Training Set was Established
As clarified in point 8, there isn't a "training set" in the AI sense for this physical device. However, the "ground truth" for the early studies (PACE, second Australian study, sheep study) that informed device development and modification was established through:
- Clinical Observation: Direct assessment by surgeons/investigators of device function, tissue expansion, patient comfort, and any adverse events or complications.
- Surgical Outcomes: Success of expansion, ability to proceed to permanent implant exchange, and any device-related failures requiring intervention.
- Animal Study Observations: Monitoring of animal subjects for pain, discomfort, and general health, along with assessment of device function in vivo.
- Device Performance Monitoring: Tracking of CO2 release, communication, and instances of under/over expansion or deflation.
These observations and outcomes from early non-clinical and clinical experience directly led to the design modifications (e.g., V2.5 with enhanced inner liner and software reprogramming) aimed at improving device performance and reducing malfunctions observed in the V2.0 version.
§ 878.3510 Carbon dioxide gas controlled tissue expander.
(a)
Identification. A carbon dioxide gas controlled tissue expander is a prescription device intended for temporary subcutaneous or submuscular implantation to stretch the skin for surgical applications, specifically to develop surgical flaps and additional tissue coverage. The device is made of an inflatable elastomer shell and is filled with carbon dioxide gas. The device utilizes a remote controller to administer doses of carbon dioxide gas from an implanted canister inside the device.(b)
Classification. Class II (special controls). The special controls for this device are:(1) In-vivo performance testing must be conducted to obtain the adverse event profile associated with use, and demonstrate that the device performs as intended under anticipated conditions of use.
(2) The patient-contacting components of the device must be demonstrated to be biocompatible.
(3) Performance data must demonstrate the sterility of patient-contacting components of the device.
(4) 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) Cycle testing of expander showing that there are no leaks or tears after repeated cycling;
(ii) Mechanical assessment of implanted carbon dioxide (CO
2 ) canister including high impact testing;(iii) Leak testing of expander showing that device does not leak CO
2 ;(iv) Assessment of gas permeability during expansion and after full expansion; and
(v) Mechanical assessment of expander (tensile set, breaking force, shell joint test, and fused or adhered joint testing).
(5) Performance data must be provided to demonstrate the electromagnetic compatibility, electrical safety, and wireless compatibility of the device.
(6) Software verification, validation, and hazard analysis must be performed.
(7) Performance data must support shelf life by demonstrating continued sterility of the device or the sterile components, package integrity, and device functionality over the identified shelf life.
(8) Human factors testing and analysis must validate that the device design and labeling are sufficient for the end user.
(9) Physician labeling must include:
(i) The operating parameters, name, and model number of the indicated external dosage controller;
(ii) Information on how the device operates and the typical course of treatment;
(iii) Information on the population for which the device has been demonstrated to be effective;
(iv) A detailed summary of the device technical parameters; and
(v) Provisions for choosing an appropriate size implant that would be exchanged for the tissue expander.
(10) Patient labeling must include:
(i) Warnings, precautions, and contraindications, and adverse events/complications;
(ii) Information on how the device operates and the typical course of treatment;
(iii) The probable risks and benefits associated with the use of the device;
(iv) Post-operative care instructions; and
(v) Alternative treatments.
(11) Patient training must include instructions for device use, when it may be necessary to contact a physician, and cautionary measures to take when the device is implanted.