K Number

Validate with FDA (Live)

Device Name

BioXmark

Manufacturer

Nanovi A/S

Date Cleared

2022-12-23

(294 days)

Product Code

Regulation Number

Type

Panel

Age Range

All

Reference & Predicate Devices

N/A

Predicate For

N/A

Intended Use

BioXmark is indicated for use to radiographically mark lung, bladder and lymph nodes in adult patients for whom it has been determined that radiographical marking of tissue for radiation treatment is indicated for their cancer treatment.

BioXmark is implanted via image guided injection into tissue relevant for radiotherapy planning at a healthcare facility. BioXmark can be implanted in the tumor, lymph nodes or tissue adjacent to the tumor subject to irradiation or healthy tissue which should not be irradiated.

BioXmark is intended to mark tissue for at least 3 months after implantation.

Device Description

The device is a single-use, implantable device consisting of a sterile transparent liquid in a onepoint-cut (OPC) glass ampoule. Upon injection of the liquid into soft tissue, efflux of alcohol leads to the formation of a radiopaque, sticky, gel-like fiducial marker in vivo, which is visible using fluoroscopy, CT. MRI, and ultrasound. The subject device is a mixture of ethanol, sucrose acetate isobutyrate (SAIB) and an iodinated and acylated derivative of sucrose (x-SAIB). Each ampoule contains 1 mL of liguid and is steam sterilized. On injection, a miniscule amount of ethanol diffuses in 1 to 2 hours, causing an increase in marker viscosity and resulting in a hydrophobic semisolid gel-like marker at the injection site.

AI/ML Overview

Here's a summary of the acceptance criteria for BioXmark and the studies performed to demonstrate that the device meets those criteria, based on the provided text:

Acceptance Criteria and Device Performance for BioXmark

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria (Special Controls & Clinical Study Endpoints)	Reported Device Performance (as demonstrated by studies)
Risk of marker migration in tissue during the course of radiation therapy through post-treatment follow-up.	Lung Cancer Study (310-01): Migration of BioXmark in lung cancer patients in this use case is less than 15% (calculated with 95% certainty). No migration observed in any of the 29 markers followed. Positional shifts were observed only in patients with large anatomical changes, with markers still at the anatomically right spot. Bladder Cancer Study: All 60 visible markers (from a total of 76 implanted) remained detectable without displacement through week four of follow-up. Blurring, migration, and fading did not occur during treatment. Animal Study (Porcine model): BioXmark markers did not migrate and retained their 3D shape over the course of a 6-week study. Animal Study (Pancreatic tumor mouse model): Position of BioXmark was stable over 4 months. Animal Study (Mice subcutaneous implantation): Followed positional stability at 1 hour, 24 hours, 1 month, and 5 months without adverse events.
Ability to visualize the marker to allow for adequate localization during the course of radiation therapy through post-treatment follow-up.	Lung Cancer Study (310-01): BioXmark was visible on all defined imaging modalities (ultrasound, CT, CBCT, 4D CT, 2D kV, fluoroscopy) during the entire treatment period. All markers had a Contrast-to-Noise Ratio (CNR) > 2 (cutoff for visibility), and CNR remained constant. Bladder Cancer Study: 79% (95% CI 70-88%) of implanted markers were visible on treatment planning CT scan. All separate spots were continuously classified as clearly visible without artifacts. All visible markers remained detectable through treatment. Bench Study (X-ray imaging): All markers (10 ul, various shapes) were clearly visible with mean contrast levels in the range of 700-1200 HU (spherical), 400-800 HU (flat cylindrical), and 500-900 HU (elongated cylindrical), resulting in CNR > 110. Bench Study (MR): All volumes of markers tested (10 ul, 25 µl, 50 µl, 100 ul) were identified with low degrees of artifacts as hypo-intense spots in both T1w FFE and T2w MRI. Animal Study (Porcine model): Visualizable using Ultrasound, fluoroscopy, and CT-imaging. Animal Study (Pancreatic tumor mouse model): Detectable by CBCT (SARRP and Skyscan).
Risk of device interference with tumor response assessment post-treatment.	Bench Study (Artifacts): Data support CNR and streaking artifacts comparable to currently marketed fiducial markers. SI of BioXmark for volumes < 200 ul (range 10.79-20.14) was significantly lower than commercial solid soft tissue markers (range 16.23-62.36) while maintaining high contrast. No relevant artifacts observed in clinical studies. Bench Study (MR): Low degree of artifacts compared to solid soft tissue markers.
All adverse events.	Lung Cancer Study (310-01): 15 SAEs and 25 AEs reported in Part A, none device-related. One AE, likely marker-related due to protocol violation (no image guidance, possible placement in a vessel/vascularized area), resulted in mild pain. 59 SAEs in Part B, no Adverse Device Effects or Serious Adverse Device Effects. Bladder Cancer Study: Two patients experienced grade 2 toxicity, one possibly related to the implantation procedure, not the marker (urinary tract infection, hematuria).
Device toxicity (Animal performance data).	Biocompatibility: Determined to be biocompatible based on cytotoxicity, sensitization, intracutaneous reactivity, acute systemic toxicity, genotoxicity, implantation, subacute/subchronic toxicity, material-mediated pyrogenicity, and toxicological risk assessment for carcinogenicity. No risk of carcinogenicity identified. Animal Safety Studies: BioXmark was well tolerated in a pancreatic tumor mouse model (no changes in physical condition or toxic side effects). No adverse events or withdrawal of animals in mice with subcutaneous implantation.
Maintenance of physical form throughout the course of therapy and post-treatment follow-up.	Lung Cancer Study (310-01): Marker volume and density were stable through the treatment period. Volume slowly reduced by ~35% over the first 9 months due to resorption, then stabilized. Contrast level remained constant for 24 months, indicating degradation from surface to interior. Animal Study (Resorption): Resorption via chemical hydrolysis, gradually degrading the surface. Degradation appears to stop at 3-9 months.
Device interference with radiation dose delivery.	Clinical Study (Dose distribution): Effect on dose distribution from BioXmark was negligible for 6 lung cancer patients treated with photon therapy. Bench Study (Proton beams): Effects appear to be equivalent or smaller than solid markers. One large BioXmark marker perturbed the proton beam by a maximum of 4.8%; no influence after the phantom. Calculated and measured Relative Proton Stopping Power (RSP) showed good agreement (1.174 calculated vs 1.164 measured).
Biocompatibility of patient-contacting components.	Biocompatibility Evaluation: Extensive testing according to ISO 10993-1 demonstrated biocompatibility.
Continued sterility, package integrity, and device functionality over the labeled shelf life.	Sterility: Sterilized by moist heat (ISO 17665-1:2006) achieving SAL of 10-6. Packaging: Shipment integrity tested per ASTM D4169-16. Shelf Life/Stability: Representative samples aged two years (real-time and accelerated) confirmed continued functionality, material, and packaging properties.
Device sterility and non-pyrogenicity.	Sterility: Demonstrated with SAL of 10-6. Pyrogenicity: Bacterial endotoxin testing per ANSI/AAMI ST72:2019.
Usability testing (device can be positioned as indicated).	Human Factors Testing: Overall success rate on critical tasks was 98.8% across 30 participants, types, and contexts (241 critical user-device interactions). Only 3 critical use errors (1.2%), attributed to study artifacts and general procedure, not specific to BioXmark.

2. Sample Size for Test Set and Data Provenance

Lung Cancer Study (310-01):
- Subjects: 15 subjects
- Markers: 29 (analyzable) out of 35 implanted markers.
- Provenance: Retrospective, collected from patients having BioXmark implanted prior to radiotherapy. The study started in February 2014 and was completed in October 2018. Country of origin not explicitly stated, but Nanovi A/S is based in Denmark.
Bladder Cancer Study:
- Subjects: 20 patients (20 evaluable for CT scan, 19 evaluable for CBCT).
- Markers: 76 implanted markers (60 evaluable for visibility on CT, 60 evaluable for stability on CBCT).
- Provenance: Prospective Phase I-II trial conducted between July 2018 and July 2019 at a tertiary university medical center. Country of origin not explicitly stated, but Nanovi A/S is based in Denmark.

3. Number of Experts and Qualifications for Test Set Ground Truth

Lung Cancer Study (310-01):
- Number of Experts: At least two (a medical physicist and an investigator).
- Qualifications: "Medical physicist" and "investigator" are stated. Specific years of experience or additional details are not provided.
Bladder Cancer Study:
- Not explicitly stated for establishing ground truth for visibility and stability. "Investigator" is mentioned in the context of the study. Visual scoring was dichotomous (visible/non-visible).

4. Adjudication Method for Test Set

Lung Cancer Study (310-01):
- Visibility was evaluated by "the naked eye by scoring yes/no by the medical physicist and investigator."
- Potential artifacts were reviewed "by the investigator and the medical physicist independently."
- For markers with decreased 3D vector length (indicating potential migration), "Independent manual assessment by both the medical physicist and the investigator resulted in their judgement that the markers had in fact not migrated." This suggests an unblinded consensus or discussion approach.
Bladder Cancer Study:
- Visibility was scored dichotomously (visible/non-visible). No explicit adjudication method described beyond this.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No MRMC comparative effectiveness study was explicitly mentioned in the provided text, nor was an effect size for human readers improving with AI vs. without AI assistance. The studies focused on the performance of the BioXmark device itself, its visibility, stability, and safety, rather than comparing human reader performance with and without AI assistance in reading images with BioXmark.

6. Standalone (Algorithm Only) Performance Study

No standalone (algorithm only) performance study was described. The performance data for BioXmark focuses on its physical properties, visibility by human assessment (medical physicist/investigator), and integration into clinical workflows.

7. Type of Ground Truth Used

Lung Cancer Study (310-01):
- Expert Consensus: For visibility, artifacts, and particularly for the re-assessment of potential migration, a joint judgment/consensus of a medical physicist and investigator was used.
- Quantitative Measurements: Contrast-to-Noise Ratio (CNR) for visibility, Hounsfield Unit (HU) thresholding for automatic contouring, and 3D vector length changes for positional stability were objective measurements.
Bladder Cancer Study:
- Visual Assessment/Expert Judgment: Visibility scored dichotomously (visible/non-visible). Positional stability scored on CT scan and latest CBCT by experts (investigators).
- Quantitative Measurements: The "preset threshold of 75%" for visibility suggests a quantitative benchmark.
Animal Studies: Primarily based on direct observation, necropsy findings, and specialized imaging techniques (CBCT, SARRP, Skyscan).
Bench Studies: Objective measurements of physical properties, attenuation coefficients, contrast levels (HU, CNR), streaking index (SI), and RSP.

8. Sample Size for the Training Set

The provided text does not describe a "training set" in the context of a machine learning algorithm. The studies mentioned (clinical, animal, bench) are performance evaluations of the physical device. Therefore, a sample size for a training set is not applicable here as no AI/ML algorithm is being 'trained'.

9. How Ground Truth for the Training Set Was Established

As there is no mention of an AI/ML algorithm or a "training set," the method for establishing ground truth for a training set is not applicable to the information provided.

Summary

{0}------------------------------------------------

DE NOVO CLASSIFICATION REQUEST FOR BIOXMARK

REGULATORY INFORMATION

FDA identifies this generic type of device as:

Phase-changing fiducial marker for radiation therapy. A phase-changing fiducial marker for radiation therapy is a single-use. sterile liquid material that changes phase in situ when injected in tissue for the purposes of aiding radiation therapy treatment. The device is intended to be visualized using one or more radiologic imaging modalities.

NEW REGULATION NUMBER: 21 CFR 892.5727

CLASSIFICATION: Class II

PRODUCT CODE: OUV

BACKGROUND

DEVICE NAME: BioXmark

SUBMISSION NUMBER: DEN220017

DATE DE NOVO RECEIVED: March 4, 2022

SPONSOR INFORMATION:

Nanovi A/S Diplomvej 378 C/O DTU Science Park Kgs. Lyngby Denmark 2800

INDICATIONS FOR USE

The BioXmark is indicated as follows:

{1}------------------------------------------------

BioXmark is intended to mark tissue for at least 3 months after implantation.

LIMITATIONS

The sale, distribution, and use of the BioXmark are restricted to prescription use in accordance with 21 CFR 801.109.

The device is sterile, single-use and cannot be reprocessed.

Do not inject the device directly into the bloodstream.

A void injections into anatomical structures such as the heart, central nervous system, necrotic tissue, and air-filled cavities.

Do not inject into patients with hypersensitivity to iodine or other components of BioXmark.

Do not exceed maximum injection volume.

PLEASE REFER TO THE LABELING FOR A COMPLETE LIST OF WARNINGS, PRECAUTIONS AND CONTRAINDICATIONS.

DEVICE DESCRIPTION

Image /page/1/Picture/11 description: The image shows a small glass ampule with a label on it. The label has the text "BioXmark" and "1 ml" printed on it. The ampule is clear and has a narrow neck. There is a small blue dot on the neck of the ampule.

Figure 1. BioXmark in single ampoule

{2}------------------------------------------------

Table 1. BioXmark composition

Liquid

Component	Description(s)	Nominal Content (w/w%)	Function
SAIB	Sucrose acetateisobutyrate	(b)(4)	Formation of marker invivo
x-SAIB	Iodinated and acylatedderivate of sucrose*		Contrast componentproviding radiopacity
EtOH	Ethanol		Solvent, reducing theviscosity of the liquid toenable injection throughthin needles (≤25G)

*Sucrose 6.6'-di(2,4,6-triiodophenoxy) isobutyrate.

Primary Packaging

Component	Description(s)	Nominal Content (w/w%)	Function
Type 1 OPC Glassampoule	Borosilicate	N/A	Sterile barrier

BioXmark is a mixture of ethanol, SAIB and x-SAIB in the ratio | 00(4) (w/w%). Upon injection, the ethanol diffuses in 1 to 2 hours, causing an increase in marker viscosity and resulting in a hydrophobic semisolid gel-like marker at the injection site. Based on performance data provided the rate of ethanol diffusion and formation of the marker is more dependent on the amount injected than the polarity of the tissue it is injected into. Ethanol diffusion rates do not vary significantly in different tissue types.

SUMMARY OF BENCH STUDIES

BIOCOMPATIBILITY

As indicated in Table 1, BioXmark is a mixture of ethanol, SAIB and x-SAIB in the overall ratio of (0(4) (w/w%). Animal testing showed evidence of limited resorption occurring for 3-9 months followed by stabilization of the marker volume. It is characterized as a permanent (>30 days) implant, tissue/bone contacting. Biocompatibility evaluation of the implant has been completed according to FDA Guidance, Use of International Standard ISO 10993-1, "Biological evaluation of medical devices - Part 1: Evaluation and testing within a risk management process." Cytotoxicity, sensitization, intracutaneous reactivity, acute systemic toxicity, genotoxicity, implantation, subacute/subchronic toxicity, and material-nediated pyrogenicity testing were conducted per the appropriate standards in the ISO-10993 series. Biocompatibility testing was performed on the sterlized device. A toxicological risk assessment was performed to evaluate the carcinogenicity endpoint for the subject device. No risk of carcinogenicity was identified. Based on all testing and evaluations, BioXmark was determined to be biocompatible.

{3}------------------------------------------------

Biocompatibility testing showed evidence of some resorption. Degradation appears to stop at the three-to-nine-month time point. The provided studies are sufficient to demonstrate lack of subchronic and chronic toxicity caused by the degradation products.

STERILITY/PACKAGING/SHELF LIFE

Sterility:

The subject device is provided sterile to the end user. The device is sterilized by moist heat in accordance with ISO 17665-1:2006. "Sterilization of health care products - Moist heat - Part 1: Requirements for the development, validation and routine control of a sterilization process for medical devices." The operation cycle selected was 121°C for 20 minutes. The sterilization cycle used for terminal sterilization of the device by moist heat achieved a F0≥15 and an SAL value of 10-6

Packaging:

The subject device is packaged in a glass ampoules are packaged in a retail box with implant cards and instructions for use.

Shipment integrity was conducted in accordance with ASTM D4169-16. "Standard Practice for Performance Testing of Shipping Containers and Systems." Packaging validation testing included stacking compression, loose load vibration, low pressure, random vibration, concentrated impact, handling, and visual inspection.

Shelf Life/Stability:

Representative sterilized samples were aged two years to determine the shelf life of the device. Testing was conducted to verify the devices still functioned as expected with two-year real time and accelerated-aged samples. Material and packaging properties did not degrade significantly during real time or accelerated aging. All tests were passed successfully.

Pyrogenicity:

Bacterial endotoxin testing performed in accordance with ANSI/AAMI ST72:2019 "Bacterial endotoxins - Test methods, routine monitoring, and alternatives to batch testing".

MAGNETIC RESONANCE (MR) COMPATIBILITY

The subject device is composed entirely of non-metallic components; per the FDA guidance document "Testing and Labeling Medical Devices for Safety in the Magnetic (MR) Environment," the device is MR Safe and poses no safety hazards in the MR environment.

HUMAN FACTORS TESTING

A human factors evaluation was conducted for the subject device in compliance with IEC 62366-1:2015+A1:2020 and FDA guidance document, "Applying Human Factors and Usability Engineering to Medical Device." Usability testing was conducted to determine if use errors associated with use of the subject device have been reduced to an acceptable level. The usability report described the validation study methods and findings that were used to evaluate the ability

{4}------------------------------------------------

of the subject device to be used as intended by the intended users in the intended use environment. Percutaneous and endoscopic implantation methods were examined, and clinical users included gastroenterologists, interventional radiologists, advanced endoscopists, and radiation oncologists, and clinical support personnel with varying levels of experience.

The validation study consisted of user performance on comprehension questions that evaluated participants' understanding of device-specific knowledge presented in the technical documentation. Simulated-use tasks were also evaluated to promote the evaluation of the BioXmark for its intended use. Usability testing was performed for both percutaneous injections and endoscopic injections.

Across the sample of thirty (30) participants, a total of 241 critical user-device interactions were scored. including 61 simulated-use tasks with discernable outcomes and 180 comprehension questions. The overall success rate on critical tasks was 98.8% across all participants, types, and contexts. In particular. the 241 critical user-device interactions vielded 238 successes (98.8%). 0 difficulties (0.0%), and only 3 use errors (1.2%). The three (3) critical use errors were all committed by the same user group, on the same task, during the simulated endoscopic implantation procedure. Importantly, all 3 of the critical use errors are attributable to study artifacts and were associated with aspects of the procedure more generally, meaning they are not specific to BioXmark. Thus, these findings demonstrate that most critical user-device interactions attempted during device use were successfully completed without any form of difficulty or use error.

PERFORMANCE TESTING - BENCH

Bench testing was conducted to demonstrate that the BioXmark performs as expected under the anticipated conditions of use. The following bench testing was conducted to demonstrate the device performance characteristics:

Test	Purpose	Results
Radiation effects on stability	To determine if radiationcauses degradation of thedevice when exposed toappropriate energy and type ofbeams	No radiation effects shown on stability.
Calculated mass attenuationcurves	To determine attenuation of thesubject device	Mass attenuation curves showed that attenuationcould be calculated based on elementalcomposition of BioXmark and X-Ray MassAttenuation Coefficients from individualelements
Effect of volume and shape	To determine of markervolume and shape affectsvisibility on X-ray imaging	All markers were visible in X-ray imaging
Artifacts	To assess marker visibility andartifact evaluation	Data provided to support CNR and streakingartifacts. Contrast levels, and streaking indexwere comparable to currently marketed fiducialmarkers. Artifacts did not impede the ability ofthe marker to be used for its intended purpose.

Table 2. Performance testing completed for BioXmark

{5}------------------------------------------------

Effects with proton beams	To measure relative proton stopping power	Effects appear to be equivalent or smaller than solid markers.
BioXmark visibility and artifacts in MR	To evaluate marker visibility and artifacts	All volumes of markers tested were identified with low degrees of artifacts.

1. Radiation effects on stability
  BioXmark markers were stress tested after exposure to a single irradiation dose of 150 Gy and compared to non-irradiated samples. HPLC analysis was performed at time: 0. ~ 30. ~60 and >90 days. Irradiation of BioXmark markers with a single dose of 150 Gy prior to storage had no effect on the hydrolysis rate. Marker degradation of non-irradiated samples was comparable to irradiated samples.

An additional study was performed, where 2.5-year-old samples were exposed to an irradiation dose of 150 Gy and compared to the non-irradiated samples. Visual inspection of BioXmark and High-performance liquid chromatography (HPLC) analysis were performed. Data showed no difference between BioXmark exposed to radiation versus the control product.

The stability of BioXmark was also studied under normo-fractionated and single-fraction proton beams to determine its stability if subjected to the conditions expected from proton beam therapy. Four markers were irradiated for a period of 51 days with 43 fractions ranging from 1.44-1.86 Gy resulting in an accumulated dose of 67.4 Gy. Four other markers were irradiated with a single dose of 155.4 Gy. BioXmark showed no signs of deterioration when exposed to proton beams.

In animal studies, absorption of the marker is seen and thought to be due to the increased inflammation following radiation therapy and increased vascularization and flow of fluid, which could lead to an increased resorption rate and a delayed formation of encapsulation typically observed in untreated animals, due to continued disturbance of the tissue. This increased resorption was not seen during the clinical testing in patients, though in some tissues, absorption was seen through months 3-9 followed by volume stabilization. Differences in resorption characterization between animal models and humans has been seen with other devices and in literature. It was concluded that this difference is not due to direct effect of radiation on the marker.

1. Calculated mass attenuation curves
  Mass attenuation coefficients as a function of photon energy were calculated for BioXmark after complete ethanol efflux, gold and soft tissue (ICRU-44) based on the elemental composition of BioXmark and X-Ray Mass Attenuation Coefficients from individual elements. Mass attenuation curves were provided. The high total iodine content of ~15 w/w% results in a high mass attenuation coefficient of the marker.
1. Effect of volume and shape on visibility under x-ray imaging

{6}------------------------------------------------

The x-ray contrast of BioXmark markers with a volume of 10 ul shaped into an elongated cylinder (rod-like), a flat cylinder (plate-like) and a sphere was investigated in a thorax phantom using CT-imaging. All markers were casted into 10 w/w% gelatin to mimic the soft tissue environment in terms of background contrast level and an anthropomorphic thorax phantom was used to mimic clinical use in patients. All markers, irrespective of 3D marker shape, were clearly visible in x-ray imaging using clinically relevant settings. A mean contrast level for the spherical BioXmark markers, flat cylindrical BioXmark markers and elongated cylindrical BioXmark markers were in the range of 700-1200 HU, 400-800 HU and 500-900 HU depending on the contouring threshold, resulting in a contrast-to-noise ratio (CNR) of > 110 in all cases.

4. Artifacts

The visibility (contrast-to-noise ratio (CNR)) and evaluation of artifacts were investigated in a thorax phantom. BioXmark was filled into hollow polypropylene spheres. CT imaging was conducted using a clinical multi slice CT scanner. Images were acquired from three different gantry angles. Exposure settings were varied. Visibility of BioXmark and the included commercial markers were based on the CNR. CBCT scans were then performed for each marker using standard exposure settings. All scans were also performed with no marker present for a baseline image with no artifact. CBCT and CT scans were used to quantify artifacts introduced by the included markers in both imaging techniques based on the streaking index (SI). The SI of BioXmark in this study was comparable to solid fiducial markers included in the study.

A second study was performed to assess the visibility and artifacts of smaller volumes of BioXmark (10 ul, 25 ul, 50 ul, 100 ul, 200 ul, 300 ul, and 400 ul). Each marker was cast into gelatin in a hollow low-density polyethylene rod container. Imaging was performed with the filled rod container placed inside a CIRS IMRT thorax phantom. CT scans were performed using a standard clinical lung protocol. The SI of BioXmark for volumes < 200 ul (range 10.79-20.14) was significantly lower than for commercial solid soft tissue markers (range 16.23-62.36) while still maintaining high contrast level as measured in HU (699-1925).

1. Effects on proton beam therapy
  The relative proton stopping power (RSP) of BioXmark was measured using a multi-layer ionization chamber to determine the water equivalent thickness. To determine the range shift of the proton beam due to BioXmark, the relative ionization of the proton beam was measured with clinical energies between 100 and 230 MeV. The water equivalent thickness for each sample was used to determine the measured RSP for BioXmark. The RSP was calculated using the chemical composition and density of BioXmark. There was good agreement between the calculated (1.174) and measured (1.164) RSP.

BioXmark markers (10 ul, 25 ul, 50 ul and 100 ul) were prepared by injecting the desired volume into distilled water to form spherical, semi-solid, gel markers were fixed in a gelatin phantom. All markers were investigated with five set-ups with different proton energies, designed to investigate the perturbation of the beam with the markers positioned at different depths relative to the Bragg Peak. For the solid markers the parallel orientation resulted in a higher dose perturbation as the protons travel through the full span of the marker material.

{7}------------------------------------------------

Only the largest of the BioXmark marker perturbed the proton beam in the spread-out Bragg Peak delivery and then a maximum perturbation of 4.8% was found in the deepest film position. No influence of the BioXmark directly after the gelatin phantom could be measured.

1. BioXmark visibility and artifacts in MR
  The visibility and introduction of possible artifacts on 3T MRI (T1w FFE and T2w) of BioXmark was investigated. The visibility and possible artifacts of BioXmark (10 ul, 25 µl, 50 µl and 100 ul) in MRI was evaluated in a pancreatic equivalent spherical MRI phantom together with two commercially available solid soft tissue markers. All volumes of BioXmark ranging from 10-100 ul were identified in both Tlw FFE and T2w MRI as hypo intense spots with low degree of artifacts compared to solid soft tissue markers.

PERFORMANCE TESTING - ANIMAL

Objectives of the following animal studies were to perform a safety and performance evaluation of the BioXmark under simulated clinical use. All objectives of the evaluations were accomplished. The device has been appropriately evaluated in pre-clinical studies. The following tests were performed:

1. In vivo kinetics of ethanol efflux
  Injection of BioXmark into tissue causes an efflux of ethanol from the marker into the adiacent tissue based on the mechanism of non-solvent induced phase separation (NIPS). The ethanol efflux kinetics from BioXmark markers in vivo were unaffected by the local tissue composition at the target injection site but dependent on the marker injection volume. The required time for complete ethanol efflux was determined to be 15-120 min post-implant. This is considered to be within the timeframe between marker placement and subsequent planning CT scan acquisition used in clinical practice, therefore the kinetics of ethanol efflux does not affect the clinical use of BioXmark.
1. In vivo resorption of BioXmark
  With ethanol efflux the material forms a highly viscous non-crystalline liquid. It was observed that resorption of the device in vivo is via chemical hydrolysis which gradually degrades the surface of the implant.
1. Visualization of BioXmark markers from chronic toxicity study
  The objective of this study was to visualize BioXmark markers following necropsy at the day of sacrifice. The rats used were from the chronic toxicity study. Subcutaneous injections at 5 injection sites in the dorsal part of the neck each administered on Day 1 of the study. BioXmark markers were visible in rats after an observation period of one year.

4. Implantation and visibility in a porcine model

The studies were conducted in order to evaluate; i) the usability of BioXmark using the identical equipment (endobronchial ultrasound (EBUS) and endoscopic

{8}------------------------------------------------

ultrasound (EUS)) currently used in the clinic operated by professional pulmonologists; ii) the visibility of BioXmark (Ultrasound-, fluoroscopy- and CTimaging) following implantation in various tissue types including lymph nodes, liver, lung, and thymus tissue. EBUS, EUS, and percutaneous injection techniques can be successfully used with the test device. Ultrasound allowed real-time monitoring of injection and the test device could be visualized using fluoroscopy after injection throughout the 6 week study. BioXmark markers did not migrate and retained their 3D shape over the course of the study.

1. Use of BioXmark for high-precision radiotherapy in a pancreatic turnor mouse model A study conducted by Dobiasch et al., 2017 established an orthotopic pancreatic tumor mouse model for high-precision IGRT using BioXmark fiducial marker. BioXmark could be detected by CBCT (SARRP and Skyscan). The position of BioXmark was monitored at least weekly by CBCT and was stable over 4 months. BioXmark was shown to be well tolerated; no changes in physical condition or toxic side effects were observed in comparison to control mice.
1. Evaluation of BioXmark for canine oral tumors for image-guided radiation therapy planning

A study conducted by Clerc-Renaud et al., 2019 evaluated BioXmark to improve identification of the superficial component of oral tumors in dogs with computed tomography imaging. This study demonstrated potential utility for combining liquid fiducial markers with post-contrast computed tomography images for improved oral tumor localization and gross tumor volumes contouring for radiation therapy planning.

1. Evaluation of BioXmark for small animal image-guided radiotherapy applications A study conducted by Brown et al., 2020 followed the positional stability of the marker at 1 hour, 24 hours, 1 month and 5 months in mice (subcutaneous implantation) without reporting any adverse events or withdrawal of animals from the study. The cumulative data supports that the subject device is safe for use with no toxicological concerns in animal subjects.

SUMMARY OF CLINICAL INFORMATION

Effect of BioXmark on dose distribution in vivo

The effect on the dose distribution from BioXmark was investigated for 6 patients treated with radiotherapy (photon therapy) for locally advanced non-small lung cancer. Two algorithms were used to evaluate the effect of dose perturbation from BioXmark. The relative dose difference in Gross Tumor Volume were calculated using both algorithms. The effect on the dose distribution from BioXmark was shown to be negligible.

The following clinical studies were provided to support the Indications for Use of the BioXmark:

A. 310-01 Lung Cancer

{9}------------------------------------------------

The Sponsor collected data from patients having BioXmark implanted prior to radiotherapy. The investigation had a Part A investigating the device in the intended use period and a Part B, a follow up period. Part A of the study was started in February 2014 and Part B was completed in October 2018.

Part A primary endpoint: Stability of the geometrical configuration and visibility of the injected gel marker in 2D kV x-rays and CBCT acquired throughout the treatment course and quantified by vectors and contrast to noise ratio (CNR). BioXmark should be visible using CT or CBCT during radiotherapy with a contrast-to-noise ratio > 2 (160 HU or more). BioXmark does not migrate during radiotherapy using a predefined procedure for capturing and measuring BioXmark marker volume, contrast and positional stability.

Part B primary endpoint: Ouantification of risk of pneumonitis and cardiotoxicity estimated by V20 lung volume, mean lung dose and V40 heart volume and late lung tissue damage for DIBH versus free breathing technique (NTCP). These endpoints are not related to BioXmark performance.

Subjects: Twenty (20) subjects were screened, eighteen (18) subjects were found suitable for inclusion and fifteen (15) subjects gave their consent. Thirty-five (35) markers were implanted in the fifteen (15) subjects, 1-3 markers (100 -300 ul) in each subject. Eleven (11) markers were placed into or near the primary tumor and twenty-four (24) into the hilar or mediastinal lymph nodes. Imaging data were not recorded for two (2) subjects due to progression of their disease (3 markers). hence for these two subjects there are no medical imaging and data are available only for assessment of procedural safety. No subjects were withdrawn or discontinued from the investigation. Two (2) markers were lost between implantation and planning CT, and one marker was not suitable as a marker for radiotherapy (placed in the wall of a tumor cavity and was dispersed). Therefore, a total of twenty-nine (29) markers were followed from implantation through radiotherapy and follow up.

Thirteen (13) subjects completed the 2-month course of radiotherapy and the injected BioXmark markers were visible on all defined imaging modalities (ultrasound, CT, CBCT, 4D CT, 2D kV and fluoroscopy). The markers were visible during the entire treatment period with no change in marker contrast level over time, irrespective of placement (healthy lung tissue, primary tumor, or within PET positive lymph nodes). CT and CBCT scans were acquired during radiotherapy and each marker was automatically contoured in the Eclipse™ contouring software by using a Hounsfield Unit thresholding function with a lower limit of 500 HU. The CT and CBCT scans and the contoured markers were subsequently analyzed for marker radiopacity, marker volume and positional stability. Visibility was evaluated by the naked eve by scoring yes/no by the medical physicist and investigator. Constant marker visibility was evident from all scans.

Marker visibility using x-ray imaging was quantitatively assessed by measuring the contrast to noise ratio (CNR). All markers had a CNR > 2, the cutoff that determines whether the marker is visible. The CNR of the markers remained constant during the entire treatment period. The recorded clinical images were reviewed for potential artifacts by the investigator and the medical physicist independently. No relevant artifacts caused by the markers were observed.

{10}------------------------------------------------

The majority of BioXmark markers showed minimal change in the 3D vector length from the Center of Volume (CoV) of the tumor/lymph node and the CoV of the marker during the entire treatment period within the pre-defined criteria for migration (<5 mm). Two markers, placed in the tumors of two (2) patients. had a maximal 3D vector length decrease of -8.4 mm and -7.8 mm at the end of treatment indicating possible marker migration. Independent manual assessment by both the medical physicist and the investigator resulted in their judgement that the markers had in fact not migrated. For one (1) patient the large decrease in 3D vector length was attributed to movement of the tumor due to significant shrinkage. The marker remained at the anatomically right spot. For the other, the surrounding tissue changed significantly due to atelectasis, the marker followed the tissue. Migration, therefore, was not observed for any of the twenty-nine (29) markers. Marker migration in lung cancer patients of ≤25% during the course of radiotherapy can be considered as a performance that is at least as good as for other soft tissue markers based on clinical experience reported in the literature. Based no observed migration with the twenty-nine (29) markers one can calculate with 95% certainty that migration of BioXmark in this use case is less than 15%.

Serious adverse events: There were fifteen (15) serious adverse events (SAEs) and twenty-five (25) adverse events reported in part A of the clinical trial, none were reported as device related. There was one adverse event that was not attributed to the device since it was the result of a protocol violation, but there is high probability that the adverse event was marker related. In violation of the protocol, the marker was placed without the use of image guidance. There is a possibility that the marker was placed in a small vessel or highly vascularized area in the tumor, and subsequently "washed out". An x-ray dense area was observed in the lateral periphery of the lung in close proximity to the pleura. The subject experienced mild pain in relation to respiration lasting less than 5 days. Two markers had been placed in the patient, therefore the patient continued in the study with the one remaining marker in situ.

There were fifty-nine (59) SAE's reported during Part B of the study. No Adverse Device Effects or Serious Adverse Device Effects or Unanticipated Serious Adverse Device Effects. were reported in Part B of Clinical Investigation #310-01 at the 36-month timepoint.

Six (6) out of the originally thirty-five (35) implanted markers were not available for analysis: three (3) because patients did not proceed to radiotherapy, two (2) because they were lost and one (1) because the marker fragmented and was not consider suitable to guide radiotherapy.

Results: BioXmark markers were suitable for image guided radiotherapy for locally advanced lung cancer patients. The marker volume and density were stable through the treatment period. The injected markers stayed within the tumor and/or lymph node position during the course of radiotherapy and were most stable for lymph node injections. Positional shifts were observed for patients with large anatomical changes. The volume of BioXmark markers was slowly reduced over the first 9 months with approximately 35% due to resorption of the marker. Further degradation could not be detected for the 9-26 month follow up period. This observation is in line with the observation from the animal studies that a fibrous capsule is created around the marker. The contrast level remained constant for the 24-month period confirming that the degradation takes part from the surface towards the interior. Migration was

{11}------------------------------------------------

assessed by visually comparing CT scans over time. Eight (8) subjects completed the 36 month follow up and fifteen (15) markers out of the thirty-five (35) originally implanted markers remained during the 36 month follow up period. There was no evidence of migration of any of the markers during the follow up period.

B. Investigator initiated study - Bladder cancer

Between July 2018 and July 2019, twenty (20) patients with non-metastasized unifocal muscle invasive bladder cancer, suitable for chemoradiation, were accrued to this prospective Phase I- II trial. The trial was conducted at a tertiary university medical center. Patients were excluded when there was any contraindication for undergoing an outpatient cystoscopy procedure.

Primary endpoints: Safety of the marker implantation procedure, marker visibility, and positional stability of the fiducial markers over time.

Predefined acceptance criteria for performance and clinical applicability were that 75% of the markers had to remain visible and positionally stable from the CT acquisition for RT planning to the last CBCT, without causing grade three toxicity (CTCAE v4.0). Formally this corresponds to test the hypothesis H0: p ≤0.75 against the alternative HA: p > 0.75. H0 is rejected if the lower limit of the 95% CI for the observed performance success in the trial is lower than 0.75.

Patients: Approximately, 3-5 BioXmark liquid markers of ~0.1 ml (0.07-0.15 ml) each were injected in proximity to the bladder tumor, using flexible cystoscopy. Fluoroscopy after injection as visual quality control measurement was used only in the first five (5) patients (19 markers) but because the implantation with the endoscopic technique was successful, further use of fluoroscopy visualization was not deemed valuable. Visibility was evaluated on radiotherapy planning CT acquisition and weekly CBCT during treatment. Visibility was scored dichotomously (visible/non-visible). Serious adverse events (SAE) associated with BioXmark marker were recorded from the moment of injection until the end of chemoradiation treatment or at least 30 days after the marker implantation. Patients were assessed weekly during treatment and 4 weeks following treatment.

In total, seventy-six (76) markers were implanted in twenty (20) patients. One patient died after CT acquisition but before start of treatment from an intercurrent disease cause. There were twenty (20) evaluable patients for marker visibility on CT scan and nineteen (19) evaluable patients for marker visibility and stability on CBCT. All nineteen (19) patients finished treatment as planned.

Of the seventy-six (76) markers implanted, sixty (60) (79% (95% CI 70-88%) were visible on treatment planning CT scan. All separate spots were continuously classified as clearly visible without artifacts. The preset threshold of 75% was overlapped by the CIs, thereby this Phase 1 study failed to show a statistical difference with the prospectively defined test. However, the study showed a clear learning curve as visibility scored 58% for the first 5 patients and 86% for the next 15. The investigator concluded that the majority of the lost markers were the result

{12}------------------------------------------------

of the implantation technique learning curve.

In patients where BioXmark was injected successfully, the fiducial marker was determined to be an easy and clinically applicable tool for IGRT in bladder-preserving chemoradiotherapy. Blurring, migration, and fading did not occur during treatment. Furthermore, this study resulted in continuous visibility of the fiducial markers. Positional stability was scored on CT scan and latest CBCT (with comparable bladder filling). Of those visible markers (60) on CT scan, all (100%) remained detectable without displacement until the end of the treatment, defined as visible on the last CBCT at week four of follow-up.

Results: BioXmark was safe to use for patients with non- metastasized unifocal muscle invasive bladder cancer. Two patients experienced grade 2 toxicity, but one of those events may be related to the implantation procedure itself and not the marker. One patient presented with urinary tract infection and the other with hematuria was present before the implantation and stopped during the radiation course.

Pediatric Extrapolation

In this De Novo request, existing clinical data were not leveraged to support the use of the device in a pediatric patient population.

LABELING

The labeling consists of Instructions for Use, device label, outer box label, and implant card.

The instructions for use include device description, storage instructions, indications for use, injection volumes and corresponding radiologic imaging modalities, warning and precaution statements, MR compatibility, shelf life, listing of adverse events, instructions for use (i.e., endoscopic assisted injections, percutaneous injection), disposal, treatment planning information and information related to patient follow-up.

The labeling meets the requirements of 21 CFR 801.109 for prescription devices.

RISKS TO HEALTH

The table below identifies the risks to health that may be associated with the use of Phasechanging fiducial markers for radiation therapy and the measures necessary to mitigate these risks:

Risks to Health	Mitigation Measures
Adverse tissue reaction	Biocompatibility evaluationAnimal performance testing

{13}------------------------------------------------

Interference with image-guidedradiation therapy or radiotherapyresponse assessment	Clinical performance testingNon-clinical performance testingLabeling
Treatment delays due to devicemalfunction, marker migration, orinability to locate marker onimaging	Clinical performance testingNon-clinical performance testingLabeling
Infection	Sterilization validationShelf life testingLabeling
Inaccurate radiation dose deliverydue to incorrect marker positioning.marker migration, or implantation	Clinical performance testingNon-clinical performance testingUsability testingLabeling
Complications due to implantationof marker or marker migration	Clinical performance testingAnimal performance testingLabeling

SPECIAL CONTROLS

In combination with the general controls of the FD&C Act, the Phase-changing fiducial marker for radiation therapy is subject to the following special controls:

Clinical performance data under anticipated conditions of use must evaluate: (1)
- (i) Risk of marker migration in tissue during the course of radiation therapy through post-treatment follow-up;
- The ability to visualize the marker to allow for adequate localization during the (ii) course of radiation therapy through post-treatment follow-up;
- (iii) Risk of device interference with tumor response assessment post-treatment; and
- All adverse events. (iv)
(2) Animal performance data under anticipated conditions of use must evaluate device toxicity and the risk of marker migration.
(3) Non-clinical performance data under anticipated conditions of use must evaluate:
- Maintenance of physical form throughout the course of therapy and post-treatment (i) follow-up;
- (ii) Device visibility on one or more radiologic imaging modalities; and
- Device interference with radiation dose delivery. (111)
(4) Performance testing must demonstrate the patient-contacting components of the device are biocompatible.
Performance testing must support the shelf life of the device by demonstrating continued (5) sterility, package integrity, and device functionality over the labeled shelf life.
Performance testing must demonstrate device sterility and non-pyrogenicity. (6)
Usability testing must demonstrate that the device can be positioned as indicated based (7) solely on reading the directions for use.
The labeling must include: (8)

{14}------------------------------------------------

(i) A detailed description of the device including materials and composition, chemical and physical properties, a description of the mechanism of the change of phase, and timeframe for achieving final state:
(ii) Summary of all reported device-related adverse events from clinical testing:
(iii) Information describing the injection procedure, including any use of image guidance, and the range of compatible injection needle gauges; and
(iv) A shelf life.

BENEFIT-RISK DETERMINATION

The BioXmark is restricted to prescription use in accordance with 21 CFR 801.109 for the radiographic marking of the lung, bladder and lymph nodes.

Device risks and mitigations:

The risks include: 1) adverse tissue reaction, 2) interference with image-guided radiation therapy or radiotherapy response assessment, 3) treatment delays due to device malfunction, marker migration or inability to locate marker on imaging, 4) infection, 5) inaccurate radiation dose delivery due to incorrect marker positioning, marker migration, or implantation, 6) complications due to implantation of marker or marker migration.

Probable benefits:

In radiation therapy, it is crucial for the physician to accurately locate the intended target volumes in order to focus radiation on the target volumes and spare healthy tissue as much as possible. Accurate target volume localization is made more challenging by organ movement and lack of exact anatomic reference points. The implantation of a fiducial marker near the tumor, or target volume, can facilitate the ease, accuracy, and precision of the localization process. However, in some tissues, implantation can be difficult, and dislocation of implanted markers is observed. As a liquid fiducial marker, BioXmark has the probability to mitigate some of the challenges observed with traditional metal fiducial markers such as imaging artifacts, technical difficulties with implantation, and marker migration or dislodgement. BioXmark is visible on several imaging modalities and creates less interference than the metal fiducial markers. Lastly, evidence has shown that in some tissue types, the marker does not migrate out of the tissue.

Evidence of benefit has only been shown in lung, bladder, and lymph nodes.

Patient Perspectives

This submission did not include specific information on patient perspectives for this device.

Benefit/Risk Conclusion

In conclusion, given the available information above, for the following indication statement:

BioXmark is implanted via image guided injection into tissue relevant for radiotherapy planning at a healthcare facility. BioXmark can be implanted in the tumor, lymph nodes

{15}------------------------------------------------

or tissue adjacent to the tumor subject to irradiation or healthy tissue which should not be irradiated.

BioXmark is intended to mark tissue for at least 3 months after implantation.

The probable benefits outweigh the probable risks for BioXmark. The device provides benefits, and the risks can be mitigated by the use of general controls and the identified special controls.

CONCLUSION

The De Novo request for the BioXmark is granted and the device is classified as follows:

Product Code: QUV Device Type: Phase-changing fiducial marker for radiation therapy Regulation Number: 21 CFR 892.5727 Class: II

Regulation Number and Section

N/A