PRE-SUBMISSION MEETING PACKAGE
QUESTIONS FOR FDA CDRH
Solstice Lumina™ Continuous Glucose Monitoring System
(Lumina C14 Implantable Sensor and Lumina R7 Wearable Reader/Transmitter)
Prepared by:
Solstice Biomedical Inc. 220 Westbrook Drive, Suite 400 Cambridge, MA 02142
With Counsel:
Calderwood & Harkness LLP 1401 K Street NW, Suite 900 Washington, D.C. 20005
Regulatory Consultant:
Pinnacle Regulatory Consulting Group LLC 9500 Medical Center Drive, Suite 210 Rockville, MD 20850
Date: June 16, 2025
CONFIDENTIAL — Contains Trade Secret and Commercial Confidential Information Subject to 21 C.F.R. § 20.61 and Exemption 4 of the Freedom of Information Act
Right-click to update Table of Contents
I. Introduction and Purpose of Questions
This section of the Pre-Submission (Q-Submission) Meeting Package sets forth specific questions that Solstice Biomedical Inc. ("Solstice" or "SBI") respectfully requests the U.S. Food and Drug Administration ("FDA"), Center for Devices and Radiological Health ("CDRH"), to address during the Pre-Submission meeting currently scheduled for August 14, 2025. The questions presented herein are intended to elicit FDA feedback on critical regulatory, clinical, and technical matters prior to the preparation and filing of a traditional 510(k) premarket notification for the Solstice Lumina™ Continuous Glucose Monitoring ("CGM") System.
Device Identification
The Solstice Lumina™ CGM System (hereinafter, the "Lumina CGM System" or the "Device") comprises two principal components designed to function as an integrated system:
1. Lumina C14 Implantable Sensor — A subcutaneously implanted, enzymatic electrochemical continuous glucose monitoring sensor with a 14-day functional wear duration. The C14 sensor employs a proprietary platinum-iridium electrode coating designated PtIr-4G.
2. Lumina R7 Wearable Reader/Transmitter — An arm-worn wearable device that wirelessly receives glucose data from the C14 sensor via Bluetooth Low Energy (BLE) 5.2 communication, processes the data through the proprietary LuminaPredict v2.1 glucose prediction algorithm, and displays real-time glucose readings and trend information to the user. The R7 reader also transmits data to the companion Lumina Mobile App (v1.0.3, available on iOS and Android platforms) for data storage, trend review, and sharing with healthcare providers.
Submission Context
Solstice intends to file a traditional 510(k) premarket notification covering both the Lumina C14 sensor and the Lumina R7 reader/transmitter as a unified system. The target submission date for the 510(k) is April 30, 2026. Solstice seeks FDA's feedback through this Pre-Submission meeting to ensure that its regulatory strategy, clinical study design, performance testing plans, and technical documentation are aligned with FDA's expectations prior to the preparation of the formal 510(k) submission.
Regulatory History
Solstice, through its U.S. subsidiary Solstice Biomedical Inc. (Delaware File No. 7834291), previously obtained 510(k) clearance for the GlucoStrip RX-200 Test Strip Reader (510(k) No. K192847, cleared November 14, 2019, product code NBW). The GlucoStrip RX-200 is a handheld blood glucose test strip reader intended for point-of-care and home use. While the Lumina CGM System represents a significant technology evolution from the GlucoStrip RX-200 — transitioning from discrete test-strip-based glucose measurement to continuous subcutaneous glucose monitoring — Solstice's prior experience with FDA submission and clearance processes informs the current regulatory strategy. The GlucoStrip RX-200 is cited herein solely for purposes of establishing Solstice's corporate pedigree and history of successful FDA interaction, and is not presented as a predicate device for substantial equivalence purposes for the Lumina CGM System.
EU Market Status
The Lumina CGM System received CE marking under the European Union Medical Device Regulation (EU) 2017/745 ("EU MDR"). The EU MDR Certificate (Certificate No. MCA-2024-07319) was issued on September 12, 2024, by Meridian Conformity Assessment GmbH, Notified Body No. 0483, a European Commission–designated Notified Body based in Munich, Germany. The Lumina CGM System has been commercially available in the European Union since September 2024, and Solstice has accumulated approximately nine months of real-world market experience in the EU as of the date of this Pre-Submission package.
Quality Management System Certification
Solstice maintains ISO 13485:2016 certification for the design, development, manufacture, and distribution of continuous glucose monitoring systems. The ISO 13485:2016 certificate was issued by Meridian Conformity Assessment GmbH and covers all manufacturing and design activities relevant to the Lumina CGM System.
Organization of Questions
The questions presented in this Pre-Submission meeting package are organized into seven (7) topic areas, as follows:
1. Predicate Acceptance and Substantial Equivalence Framework
2. Acceptability of Proposed Clinical Study Design and Endpoints
3. IDE Exemption Applicability for Non-Significant Risk Study
4. Software Documentation Requirements and Level of Concern Designation
5. Cybersecurity Premarket Submission Expectations
6. Biocompatibility Testing Scope Adequacy
7. iCGM Classification Pathway Considerations
For each topic area, Solstice provides brief contextual information to frame the question(s), followed by the specific question(s) for FDA. A consolidated summary table of all questions is provided in Section IX for ease of reference during the Pre-Submission meeting.
II. Topic 1: Predicate Acceptance and Substantial Equivalence Framework
Context
Solstice has undertaken a thorough review of legally marketed devices in order to identify appropriate predicate devices for purposes of establishing substantial equivalence of the Lumina CGM System under Section 510(k) of the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 360(k). Based on this review, Solstice has identified the following predicate devices:
Primary Predicate Device
NovaSense GlucoTrack NX-500 CGM System
The NovaSense GlucoTrack NX-500 was selected as the primary predicate because it shares the same product code (QKQ), classification regulation (21 C.F.R. § 862.1355), intended use (continuous glucose monitoring in persons with diabetes mellitus), and fundamental technology platform (enzymatic electrochemical subcutaneous sensor with wireless data transmission to a dedicated reader device) as the Lumina CGM System.
Secondary Predicate (Reference Device)
NovaSense GlucoTrack NX-700 CGM System
The NovaSense GlucoTrack NX-700 is cited as a secondary predicate (reference device) to bridge the difference in sensor wear duration between the Lumina CGM System (14 days) and the primary predicate NX-500 (10 days). The NX-700, which was cleared approximately two years after the NX-500, extends the sensor wear duration to 14 days — the same duration as the Lumina C14 sensor. Additionally, the NX-700 employs an enhanced BLE 5.0 wireless communication protocol, which partially bridges the wireless communication difference between the Lumina CGM System (BLE 5.2) and the primary predicate NX-500 (BLE, version not specified as 5.0 in the original NX-500 clearance materials but understood to be an earlier BLE version).
Additional Reference Device (Corporate Pedigree Only)
Solstice GlucoStrip RX-200 Test Strip Reader
As noted in Section I, the GlucoStrip RX-200 is cited solely for purposes of establishing Solstice's prior regulatory history and corporate pedigree of FDA interaction. It is not presented as a predicate device for the Lumina CGM System and is not relied upon for any element of the substantial equivalence argument.
Key Technological Differences Between the Lumina CGM System and the Primary Predicate (K211034)
Solstice has identified the following five principal technological differences between the Lumina CGM System and the primary predicate NovaSense GlucoTrack NX-500 (K211034):
1. Sensor Wear Duration: The Lumina C14 sensor is designed for 14-day continuous wear, compared to the 10-day wear duration of the NX-500 sensor. Solstice notes that this difference is bridged by the secondary predicate NX-700 (K230489), which also features a 14-day sensor wear duration.
2. Electrode Material: The Lumina C14 sensor employs a proprietary platinum-iridium electrode with a PtIr-4G coating, whereas the NX-500 sensor uses a platinum electrode. The PtIr-4G coating was developed by Solstice to enhance sensor stability and reduce signal drift over the extended 14-day wear period. While the base electrode material (platinum) is shared between the devices, the addition of iridium and the proprietary PtIr-4G coating represents a material difference that Solstice intends to address through biocompatibility testing and bench performance data.
3. BLE Version: The Lumina CGM System uses Bluetooth Low Energy version 5.2 for wireless communication between the C14 sensor and the R7 reader, and between the R7 reader and the companion mobile application. The NX-500 uses an earlier version of BLE (the NX-700 secondary predicate uses BLE 5.0). BLE 5.2 offers enhanced data throughput, improved connection stability, and additional security features compared to earlier BLE versions, but the fundamental wireless communication modality (BLE) remains the same.
4. Reader Form Factor: The Lumina R7 reader/transmitter is an arm-worn wearable device, whereas the NX-500 employs a handheld receiver. The change in form factor is intended to improve user convenience and enable continuous, hands-free glucose monitoring. The core functionality of the reader — receiving sensor data, processing the data through the glucose algorithm, and displaying glucose readings — remains substantially equivalent regardless of form factor.
5. Algorithm: The Lumina CGM System employs the Solstice proprietary glucose prediction algorithm designated LuminaPredict v2.1, which differs from the NovaSense algorithm used in both predicate devices (NX-500 and NX-700). The LuminaPredict v2.1 algorithm was developed by Solstice's internal software and data science teams and has been validated against both laboratory glucose reference data and clinical data generated during the EU MDR conformity assessment process.
Solstice's Position on Substantial Equivalence
It is Solstice's position that the five identified technological differences between the Lumina CGM System and the primary predicate (K211034) do not raise new questions of safety or effectiveness. Solstice believes that performance testing data (including bench testing, biocompatibility testing, software verification and validation, and electromagnetic compatibility testing) and clinical data from the proposed LUMINA-US pivotal clinical study will collectively demonstrate that the Lumina CGM System is as safe and effective as the primary predicate, and therefore substantially equivalent within the meaning of Section 513(i) of the Federal Food, Drug, and Cosmetic Act, 21 U.S.C. § 360c(i).
A detailed predicate comparison matrix and substantial equivalence discussion are provided in the companion section of this Pre-Submission meeting package entitled Predicate Comparison and Substantial Equivalence Discussion.
Questions
Question 1.1: Does FDA concur that the NovaSense GlucoTrack NX-500 CGM System (510(k) No. K211034) is an appropriate primary predicate device for the Solstice Lumina™ CGM System, given that both devices share the same product code (QKQ), classification regulation (21 C.F.R. § 862.1355), and intended use for continuous glucose monitoring in persons with diabetes mellitus?
Question 1.2: Does FDA concur that the NovaSense GlucoTrack NX-700 CGM System (510(k) No. K230489) may be appropriately cited as a secondary predicate (reference device) to bridge the difference in sensor wear duration (14 days vs. 10 days for the primary predicate)?
Question 1.3: Does FDA agree that the five identified technological differences between the Lumina CGM System and the primary predicate — specifically: (i) sensor wear duration, (ii) electrode material (platinum-iridium with PtIr-4G coating vs. platinum), (iii) BLE version (5.2 vs. earlier BLE), (iv) reader form factor (arm-worn wearable vs. handheld receiver), and (v) algorithm (LuminaPredict v2.1 vs. NovaSense algorithm) — can be adequately addressed through bench performance testing and a prospective clinical study, without requiring a De Novo classification request or Premarket Approval Application (PMA)?
Question 1.4: Does FDA have any concerns with the substantial equivalence framework as presented in the accompanying Predicate Comparison and Substantial Equivalence Discussion section of this Pre-Sub package? If so, Solstice respectfully requests that FDA identify any specific areas of concern so that they may be addressed in the 510(k) submission.
III. Topic 2: Acceptability of Proposed Clinical Study Design and Endpoints
Context
Solstice proposes to conduct a pivotal clinical study to generate the clinical data necessary to support the 510(k) submission for the Lumina CGM System. The study, designated Protocol SBD-LUM-CL-001 and titled "LUMINA-US," is described in summary below. A detailed protocol synopsis is provided in the companion section of this Pre-Submission meeting package entitled Proposed Clinical Study Protocol Summary.
Study Design. The LUMINA-US study is designed as a prospective, single-arm, multicenter clinical study to evaluate the accuracy, safety, and performance of the Solstice Lumina™ CGM System in adult subjects with Type 1 diabetes mellitus (T1D) and Type 2 diabetes mellitus (T2D).
Enrollment and Study Population
Solstice proposes to enroll a total of 350 subjects, stratified as follows:
• 175 subjects with Type 1 diabetes mellitus (T1D)
• 175 subjects with Type 2 diabetes mellitus (T2D)
All subjects will be aged 18 years and older. The equal stratification between T1D and T2D is intended to ensure that the study population is representative of the intended use population and to permit subgroup analysis of sensor performance across diabetes types.
Clinical Sites. The study will be conducted at six (6) U.S. clinical sites. Site selection is ongoing, and Solstice anticipates that sites will be distributed across geographically diverse regions to enhance the generalizability of study findings.
Contract Research Organization
The clinical study will be managed by Veltrano Clinical Research Organization Inc., 500 North Capitol Street, Suite 700, Indianapolis, IN 46204. Veltrano is an experienced CRO with a track record of managing FDA-regulated medical device clinical studies, including prior CGM studies.
Institutional Review Board
The study will be overseen by Clearwater Institutional Review Board LLC, 3200 Bayshore Boulevard, Tampa, FL 33629 (IRB Registration No. IRB-2019-04821). Clearwater IRB is a commercial IRB registered with the Office for Human Research Protections (OHRP) and accredited by the Association for the Accreditation of Human Research Protection Programs (AAHRPP).
Primary Endpoint. The primary study endpoint is sensor accuracy as measured by the Mean Absolute Relative Difference (MARD) between paired glucose measurements from the Lumina CGM System and a laboratory reference glucose analyzer. Solstice has established a performance target of MARD ≤ 9.0% overall, which is consistent with FDA's current expectations for integrated continuous glucose monitoring (iCGM) classification as articulated in the FDA guidance document Integrated Continuous Glucose Monitoring Systems (September 2023).
Secondary Endpoints. The following secondary endpoints will be evaluated:
• Consensus Error Grid Analysis: Target of ≥ 99% of paired data points falling within Zones A and B of the Clarke Error Grid Analysis. Solstice notes that the use of the Clarke Error Grid is consistent with prior CGM 510(k) submissions and FDA guidance, but seeks FDA's input on whether alternative error grid methodologies (e.g., Surveillance Error Grid) are preferred or required.
• Safety Endpoints: Adverse event rates, including but not limited to: device-related skin reactions (erythema, edema, induration), infection at the sensor insertion site, sensor fracture or breakage, and any serious adverse events (SAEs).
• User Labeling Comprehension: Evaluation of user understanding of key labeling elements, including sensor insertion instructions, glucose reading interpretation, and alert/alarm response, through a structured labeling comprehension assessment.
Additional Performance Parameters
In addition to the primary and secondary endpoints, the LUMINA-US study will evaluate the following performance parameters:
Estimated Study Duration and Timeline
The estimated total study duration of approximately six months encompasses 14 weeks for enrollment, a 14-day sensor wear period per subject, and approximately 8 weeks for data cleaning, analysis, and clinical study report preparation. The planned study start of September 15, 2025, is subject to receipt of Pre-Submission feedback from FDA and final IRB approval by Clearwater IRB.
Study Budget. The total estimated budget for the LUMINA-US clinical study is $4,200,000, based on an estimated per-subject cost of approximately $12,000 across 350 enrolled subjects. This budget encompasses clinical site fees, CRO management costs, IRB fees, device costs, laboratory reference glucose analyses, data management, biostatistical analysis, and clinical study report preparation.
Questions
Question 2.1: Does FDA concur that the proposed clinical study design — specifically, a prospective, single-arm, multicenter study enrolling 350 subjects (175 T1D and 175 T2D) at 6 U.S. clinical sites — is acceptable to support a 510(k) submission for the Lumina CGM System? If FDA has recommendations regarding alternative study designs (e.g., randomized, controlled, or comparative designs), Solstice respectfully requests FDA's guidance.
Question 2.2: Does FDA agree that the primary endpoint of MARD ≤ 9.0% overall is an appropriate accuracy metric for the proposed clinical study? Additionally, does FDA have any recommendations regarding stratified accuracy thresholds — for example, accuracy targets stratified by glucose range (e.g., hypoglycemic range < 70 mg/dL, euglycemic range 70–180 mg/dL, hyperglycemic range > 180 mg/dL), by diabetes type (T1D vs. T2D), or by day of sensor wear (e.g., days 1–7 vs. days 8–14)?
Question 2.3: Does FDA concur that the Clarke Error Grid Analysis with a target of ≥ 99% of paired data points in Zones A+B is an acceptable consensus error grid analysis methodology for the LUMINA-US study, or does FDA prefer or require the use of the Surveillance Error Grid (SEG) or another alternative error grid methodology? If alternative methodologies are preferred, does FDA have specific performance targets or acceptance criteria that Solstice should apply?
Question 2.4: Does FDA agree that the proposed sample size of 350 subjects (175 T1D + 175 T2D) is adequate to demonstrate sensor accuracy and safety with sufficient statistical power, or does FDA recommend a different enrollment target or different stratification? Solstice notes that the proposed sample size is intended to provide adequate statistical power for both overall MARD analysis and subgroup analyses by diabetes type.
Question 2.5: Does FDA have specific recommendations regarding: (i) the number and type of clinical sites (e.g., academic medical centers vs. community diabetes clinics); (ii) the reference blood glucose measurement methodology (e.g., YSI laboratory analyzer, hexokinase method, or other reference methods); or (iii) the frequency and timing of paired comparator measurements during the 14-day sensor wear period (e.g., minimum number of paired points per subject, distribution of measurements across the glucose range)?
Question 2.6: Does FDA concur that user labeling comprehension testing may be incorporated into the pivotal clinical study as a secondary endpoint, or does FDA recommend that Solstice conduct a separate, standalone human factors validation study? If a separate study is recommended, does FDA have expectations regarding the design, population, or scope of such a study?
IV. Topic 3: IDE Exemption Applicability for Non-Significant Risk Study
Context
Solstice anticipates that the LUMINA-US clinical study (Protocol SBD-LUM-CL-001) will qualify for the Investigational Device Exemption (IDE) exemption applicable to non-significant risk (NSR) diagnostic device studies under 21 C.F.R. § 812.2(c)(3). Under this provision, a clinical study of a diagnostic device may be exempt from the full IDE requirements set forth in 21 C.F.R. Part 812, provided that the study is determined to present non-significant risk and is conducted under IRB oversight.
Rationale for NSR Determination
Solstice has assessed the risk profile of the LUMINA-US study and has determined that the study presents non-significant risk for the following reasons:
1. Diagnostic Device: The Lumina CGM System is a diagnostic device intended to continuously measure interstitial glucose concentrations and display glucose readings to the user. The device does not directly administer therapy, control drug delivery, or actuate any therapeutic intervention. The device is not an automated insulin dosing controller and does not autonomously adjust insulin delivery.
2. Superficial Implantation: The Lumina C14 sensor is implanted subcutaneously (i.e., in the subcutaneous tissue beneath the skin) at a shallow insertion depth. The implantation procedure is minimally invasive, performed by the user via a spring-loaded applicator, and does not involve deep tissue, organ, or vascular implantation. The sensor is designed to be easily removed by the user at the end of the 14-day wear period.
3. Limited Duration of Implantation: The C14 sensor wear duration is 14 days (336 hours), after which the sensor is removed and discarded. This duration falls within the range of existing legally marketed CGM sensors that have been studied under NSR protocols.
4. Historical Precedent: Similar CGM clinical studies, including studies of subcutaneously implanted enzymatic electrochemical glucose sensors with comparable or longer sensor wear durations, have historically been determined to be non-significant risk studies. Solstice is aware that multiple prior CGM 510(k) submissions have been supported by clinical data generated under NSR study protocols with IRB oversight.
5. No Adjunctive Use Limitation Removed: The LUMINA-US study protocol does not direct subjects to make insulin dosing decisions based solely on the Lumina CGM System readings. Subjects will be instructed to confirm glucose readings with a blood glucose meter before making treatment decisions, consistent with standard CGM study protocols.
IRB Oversight. Under the NSR determination, the LUMINA-US study would proceed under abbreviated IDE requirements with oversight by Clearwater Institutional Review Board LLC (IRB Registration No. IRB-2019-04821), without requiring a separate IDE application to FDA. Clearwater IRB would review and approve the study protocol, informed consent documents, and study conduct in accordance with 21 C.F.R. Parts 50, 56, and 812.
Solstice seeks FDA's input on the appropriateness of the NSR designation prior to initiating the study to ensure alignment with FDA's expectations and to avoid the need for a subsequent IDE submission that could delay the study timeline.
Questions
Question 3.1: Does FDA agree that the proposed LUMINA-US clinical study (Protocol SBD-LUM-CL-001) qualifies as a non-significant risk (NSR) diagnostic device study under 21 C.F.R. § 812.2(c)(3), and therefore may proceed under the abbreviated IDE requirements with IRB approval by Clearwater Institutional Review Board LLC (IRB Registration No. IRB-2019-04821) rather than requiring a separate IDE application to FDA?
Question 3.2: If FDA considers the LUMINA-US study to be a significant risk (SR) study, Solstice respectfully requests that FDA identify the specific factors driving that determination. Additionally, Solstice requests FDA's guidance on what specific elements and documentation would need to be included in a full IDE submission in order to initiate the study in a timely manner consistent with Solstice's planned study start date of September 15, 2025.
Question 3.3: Does the 14-day subcutaneous implantation duration of the Lumina C14 sensor affect the NSR/SR determination in FDA's assessment? Solstice notes that some CGM devices with similar or longer sensor wear durations (e.g., 14-day and 15-day subcutaneous sensors) have been studied under NSR protocols, and seeks confirmation that the 14-day wear duration does not independently elevate the risk classification of the study.
V. Topic 4: Software Documentation Requirements and Level of Concern Designation
Context
The Lumina CGM System incorporates multiple software components that are integral to the device's intended function. These software components collectively govern sensor data acquisition, signal processing, glucose value computation, data display, wireless communication, and data storage. Each software component has been developed and is maintained in accordance with the requirements of IEC 62304:2006/A1:2015 (Medical device software — Software life cycle processes) and IEC 62366-1:2015 (Medical devices — Part 1: Application of usability engineering to medical devices).
Software Components. The Lumina CGM System includes the following four software components:
1. Lumina Firmware v3.4.2 — Embedded software resident on the Lumina C14 sensor. This firmware controls sensor electrode excitation, analog-to-digital conversion of sensor signals, initial signal processing, and BLE data transmission from the sensor to the R7 reader. The firmware operates on a low-power microcontroller within the sensor assembly.
2. Lumina Reader OS v2.1.0 — Operating software resident on the Lumina R7 wearable reader/transmitter. This software manages BLE communication with the C14 sensor, receives and processes raw sensor data, applies the LuminaPredict glucose prediction algorithm, generates glucose readings and trend displays, manages alert and alarm functions, and communicates with the companion Lumina Mobile App via BLE.
3. LuminaPredict Algorithm v2.1 — A proprietary glucose prediction algorithm developed by Solstice's data science team. The algorithm receives processed sensor signals from the R7 reader OS, applies calibration models and signal correction routines, and outputs real-time glucose concentration values and glucose trend predictions. The algorithm was developed using machine learning techniques applied to a training data set comprising glucose sensor data paired with laboratory reference glucose measurements obtained during the EU MDR clinical study and prior feasibility studies. The algorithm has been validated against independent validation data sets not used during training.
4. Lumina Mobile App v1.0.3 — A companion mobile application available on iOS and Android platforms. The mobile application receives glucose data from the R7 reader via BLE, displays real-time and historical glucose readings, generates glucose trend reports, and enables data sharing with healthcare providers via a secure cloud-based data platform. The mobile application does not independently generate glucose readings and is dependent upon the R7 reader for all glucose computation.
Level of Concern Designation
Solstice has classified the overall software level of concern for the Lumina CGM System as "Major" based on the framework described in the FDA guidance document Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices and consistent with the software safety classification requirements of IEC 62304:2006/A1:2015. The "Major" level of concern designation reflects Solstice's assessment that a software failure or defect in the Lumina CGM System could result in incorrect glucose readings that, if relied upon for treatment decisions, could lead to serious injury (e.g., hypoglycemia or hyperglycemia) but would not directly result in death in the absence of other mitigating factors (e.g., the device labeling instructs users to confirm readings with a blood glucose meter before making treatment decisions).
Solstice notes that the "Major" designation applies to the software components that are directly involved in glucose measurement and display (i.e., the sensor firmware, reader OS, and LuminaPredict algorithm). Solstice has assessed the companion mobile application as a lower-risk software component, given that it does not independently generate glucose readings but merely receives and displays data computed by the R7 reader. However, Solstice seeks FDA's input on whether the mobile application warrants the same level of documentation as the primary software components.
Software Documentation. Solstice has developed comprehensive software lifecycle documentation in accordance with IEC 62304:2006/A1:2015 for all four software components. This documentation includes, but is not limited to: software development plans, software requirements specifications, software architecture documents, detailed design documents, unit test plans and reports, integration test plans and reports, system test plans and reports, software release documentation, configuration management documentation, and software problem resolution documentation.
Algorithm Validation. The LuminaPredict v2.1 algorithm is a new proprietary algorithm that differs from the NovaSense algorithm used in the predicate devices (K211034 and K230489). Solstice recognizes that this algorithmic difference may warrant additional scrutiny from FDA. Solstice has conducted extensive algorithm validation activities, including: (i) retrospective performance analysis using data from the EU MDR clinical study; (ii) prospective verification using data from a U.S.-based feasibility study (n=50 subjects); (iii) sensitivity analysis across glucose ranges; and (iv) stress testing under simulated adverse signal conditions (e.g., sensor drift, signal noise, interference). Solstice intends to include a summary of algorithm validation data in the 510(k) submission and seeks FDA's guidance on the scope and depth of algorithm documentation expected.
Questions
Question 4.1: Does FDA concur with the designation of the Lumina CGM System software as "Major" level of concern? Does FDA have any recommendations regarding the appropriate software documentation scope and depth for the 510(k) submission, including whether the "Major" level of concern designation should apply uniformly to all four software components or whether a differentiated approach (e.g., "Major" for sensor firmware, reader OS, and algorithm; "Moderate" for the mobile application) would be acceptable?
Question 4.2: Given that the LuminaPredict v2.1 algorithm is a new proprietary algorithm differing from the predicate device algorithm, does FDA require specific algorithm validation documentation beyond what would typically be included in a 510(k) submission for a CGM system? For example, does FDA expect: (i) a detailed algorithmic description (e.g., signal processing pipeline, calibration model structure, machine learning methodology); (ii) a description of training and validation data sets, including data set size, composition, and representativeness; (iii) retrospective performance analysis on independent validation data; or (iv) any other algorithm-specific documentation?
Question 4.3: Does FDA have specific expectations regarding the software documentation for the companion Lumina Mobile App v1.0.3? In particular: (i) should the mobile application be addressed as a separate device (e.g., as a Software as a Medical Device (SaMD)) or as a component of the Lumina CGM System within the same 510(k) submission; (ii) does the mobile application require its own level of concern classification; and (iii) does FDA have expectations regarding the depth of software lifecycle documentation for the mobile application?
Question 4.4: Does FDA require a complete IEC 62304 software lifecycle documentation package — including software development plan, software requirements specification, software architecture document, and unit/integration/system testing documentation — for all four software components (sensor firmware, reader OS, algorithm, and mobile app)? Alternatively, are there areas where FDA would accept a risk-based, streamlined documentation approach — for example, submission of hazard analyses, key test summaries, and traceability matrices in lieu of complete lifecycle documents for lower-risk software components?
VI. Topic 5: Cybersecurity Premarket Submission Expectations
Context
The Lumina CGM System employs wireless communication technologies and cloud-based data infrastructure that necessitate robust cybersecurity considerations. Solstice has developed a comprehensive cybersecurity program for the Lumina CGM System and seeks FDA's input on specific cybersecurity documentation expectations for the 510(k) submission.
Wireless Communication Architecture
The Lumina CGM System uses Bluetooth Low Energy (BLE) 5.2 as the wireless communication protocol for two primary data links:
• Link 1: C14 Sensor → R7 Reader/Transmitter (BLE 5.2, encrypted, proximity-paired)
• Link 2: R7 Reader/Transmitter → Lumina Mobile App (BLE 5.2, authenticated pairing)
Additionally, the Lumina Mobile App connects to Solstice Cloud Services (hosted on Ridgeway Industries (Meridian Technologies) infrastructure in the U.S. East region) for data storage, data aggregation, and data sharing with healthcare providers. The cloud connection utilizes TLS 1.3 encryption for all data in transit and AES-256 encryption for data at rest.
Threat Modeling. Solstice has completed a comprehensive threat modeling exercise for the Lumina CGM System using the STRIDE (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, Elevation of Privilege) framework. The threat model identifies potential cybersecurity threats across all data links and system interfaces, assesses the likelihood and impact of each threat, and documents the mitigation controls implemented to address each identified threat. The threat model is maintained as a living document and is updated as part of Solstice's ongoing software development and maintenance processes.
Software Bill of Materials
Solstice intends to include a Software Bill of Materials (SBOM) in the 510(k) submission. The SBOM will identify all software components, including proprietary software developed by Solstice and third-party software libraries and components incorporated into the device software. Solstice seeks FDA's input on the expected scope and format of the SBOM.
Vulnerability Assessment and Management
Solstice has conducted vulnerability assessments of all software components and system interfaces, including static code analysis, dynamic testing, and penetration testing. Solstice has also developed a vulnerability management process that includes ongoing monitoring for newly discovered vulnerabilities in third-party software components, risk assessment of identified vulnerabilities, and a defined process for developing and deploying software patches and updates.
Software Update and Patch Management
Solstice has developed a software update and patch management plan that addresses the process for identifying, developing, testing, validating, and deploying software updates and security patches to the Lumina CGM System throughout the product lifecycle. The plan addresses both planned updates (e.g., feature enhancements, performance improvements) and unplanned updates (e.g., security patches in response to newly discovered vulnerabilities).
Authentication and Encryption Protocols
The Lumina CGM System implements the following authentication and encryption protocols:
• BLE 5.2 Pairing: Secure Connections pairing with LE Secure Connections (LESC) using Elliptic Curve Diffie-Hellman (ECDH) key exchange
• Data Encryption (BLE): AES-CCM 128-bit encryption for all BLE data transmissions
• Data Encryption (Cloud): TLS 1.3 for data in transit; AES-256 for data at rest
• User Authentication (Mobile App): Multi-factor authentication (MFA) for cloud account access
• Device Authentication: Mutual authentication between C14 sensor and R7 reader using certificate-based authentication
Applicable FDA Guidance
Solstice is addressing cybersecurity considerations in accordance with the FDA guidance document Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions (September 2023). Solstice's cybersecurity documentation is structured to align with the content recommendations in this guidance, including the threat model, cybersecurity risk assessment, SBOM, vulnerability testing, software update plan, and cybersecurity controls documentation.
Questions
Question 5.1: Does FDA concur that the cybersecurity documentation framework described in FDA's September 2023 guidance (Cybersecurity in Medical Devices: Quality System Considerations and Content of Premarket Submissions) is the appropriate framework for the Lumina CGM System 510(k) submission? Additionally, does FDA have any device-specific cybersecurity expectations beyond the general guidance that would be applicable to CGM systems, given the wireless communication architecture and cloud connectivity of the Lumina CGM System?
Question 5.2: Does FDA have specific expectations regarding the scope and format of the Software Bill of Materials (SBOM) for the Lumina CGM System? Specifically: (i) must the SBOM cover all four software components (sensor firmware, reader OS, prediction algorithm, and mobile app); (ii) must the SBOM include all third-party software libraries and components, regardless of whether they are open-source or commercially licensed; and (iii) does FDA have a preferred SBOM format (e.g., SPDX, CycloneDX, or other format)?
Question 5.3: Does FDA require documentation of a specific software update and patch management plan as part of the premarket 510(k) submission? If so, what level of detail is expected — for example, does FDA expect a high-level policy document, a detailed procedural document, or both? Does FDA have expectations regarding the defined timeframes for addressing critical vulnerabilities versus non-critical vulnerabilities?
Question 5.4: Does FDA have specific expectations regarding the encryption and authentication protocols for BLE 5.2 wireless communication between the C14 sensor and R7 reader, and between the R7 reader and the mobile application? In particular, Solstice notes that BLE 5.2 represents an updated wireless protocol compared to the BLE 5.0 used in the secondary predicate NX-700 (K230489), and seeks confirmation that the BLE 5.2 security features (including LESC pairing and AES-CCM 128-bit encryption) meet FDA's cybersecurity expectations for a CGM system. Does FDA have any concerns about the wireless communication security architecture as described above?
VII. Topic 6: Biocompatibility Testing Scope Adequacy
Context
Biocompatibility testing for the Lumina CGM System is being conducted in accordance with ISO 10993-1:2018 (Biological evaluation of medical devices — Part 1: Evaluation and review within a risk management process). The biocompatibility evaluation follows a risk-based approach, considering the nature of patient contact, the duration of contact, and the specific materials of construction for each device component.
Contract Testing Laboratory
All biocompatibility testing is being performed by Whitmore Laboratories Inc., 1625 Shady Grove Road, Suite 300, Gaithersburg, MD 20878. Whitmore Laboratories holds A2LA (American Association for Laboratory Accreditation) accreditation (Certificate No. 3847.01) and is compliant with ISO/IEC 17025:2017 (General requirements for the competence of testing and calibration laboratories). Whitmore Laboratories is an FDA-recognized contract testing laboratory with extensive experience in medical device biocompatibility testing, including ISO 10993 testing for implantable and skin-contact devices.
Biocompatibility Testing Scope — Lumina C14 Sensor
The Lumina C14 sensor is a subcutaneously implanted device with a maximum contact duration of 14 days (≤ 30 days). Under the ISO 10993-1:2018 framework, the C14 sensor is classified as an implant device with limited contact duration (≤ 30 days), requiring evaluation for the following biological endpoints:
This testing scope encompasses all biological endpoints recommended by ISO 10993-1:2018 Table A.1 for an implant device with limited contact duration.
Biocompatibility Testing Scope — Lumina R7 Reader
The Lumina R7 reader/transmitter is an external, skin-contacting device (arm-worn wearable) with a contact duration classified as prolonged (> 24 hours to ≤ 30 days). Under the ISO 10993-1:2018 framework, the R7 reader is classified as a surface-contacting device (skin contact) with prolonged contact duration, requiring evaluation for the following biological endpoints:
Proprietary PtIr-4G Electrode Coating
Solstice specifically notes that the Lumina C14 sensor employs a proprietary platinum-iridium electrode coating designated PtIr-4G. This coating technology was developed by Solstice to enhance sensor signal stability, reduce interfering substance sensitivity, and improve long-term sensor accuracy over the 14-day wear duration. The PtIr-4G coating represents a new material that was not used in the primary predicate device (NX-500, K211034), which employs a platinum electrode, or in the secondary predicate device (NX-700, K230489).
Solstice recognizes that the introduction of a new electrode coating material may warrant additional biocompatibility evaluation beyond the standard ISO 10993-1 test battery. Solstice seeks FDA's guidance on whether additional material characterization studies (e.g., chemical characterization per ISO 10993-18, extractables and leachables studies) are required for the PtIr-4G coating.
Target Completion Date
All biocompatibility testing for both the C14 sensor and R7 reader is targeted for completion by August 31, 2025. Final biocompatibility reports from Whitmore Laboratories are expected to be available by approximately October 2025, providing sufficient lead time for inclusion in the 510(k) submission targeted for April 30, 2026.
Questions
Question 6.1: Does FDA concur that the proposed biocompatibility testing scope for the Lumina C14 sensor — specifically, cytotoxicity (ISO 10993-5), sensitization (ISO 10993-10), irritation (ISO 10993-23), acute systemic toxicity (ISO 10993-11), subchronic toxicity (ISO 10993-11), implantation (ISO 10993-6), hemocompatibility (ISO 10993-4), and genotoxicity (ISO 10993-3) — is adequate for a subcutaneously implanted sensor with up to 14-day contact duration? Does FDA recommend any additional biological endpoints not included in the proposed testing scope?
Question 6.2: Given that the C14 sensor uses a proprietary platinum-iridium (PtIr-4G) electrode coating that differs from the platinum electrode used in the primary predicate device (K211034), does FDA require any additional biocompatibility endpoints or material characterization studies specific to the PtIr-4G coating? For example, does FDA expect: (i) chemical characterization per ISO 10993-18 (Chemical characterization of medical device materials within a risk management process); (ii) extractables and leachables studies to characterize substances that may be released from the PtIr-4G coating during the 14-day subcutaneous implantation period; or (iii) any other material-specific evaluation?
Question 6.3: Does FDA concur that the proposed biocompatibility testing scope for the Lumina R7 reader — specifically, cytotoxicity (ISO 10993-5), sensitization (ISO 10993-10), and irritation (ISO 10993-23) — is adequate for a skin-contact wearable device with prolonged contact duration? Does FDA recommend any additional biological endpoints for the R7 reader?
Question 6.4: Does FDA accept biocompatibility testing data generated at an A2LA-accredited, ISO 17025-compliant contract testing laboratory (Whitmore Laboratories Inc., A2LA Certificate No. 3847.01)? Are there any specific laboratory qualification, accreditation, or FDA recognition requirements that Solstice should be aware of that may affect the acceptability of the biocompatibility testing data in the 510(k) submission?
VIII. Topic 7: iCGM Classification Pathway Considerations
Context
In 2018, FDA established the integrated Continuous Glucose Monitoring (iCGM) device classification through De Novo classification order DEN170088, creating a new regulatory classification for CGM devices that meet specified performance criteria and are designed to be interoperable with other compatible medical devices, such as automated insulin dosing (AID) systems and insulin pumps. The iCGM classification carries the product code QKQ and is regulated under 21 C.F.R. § 862.1355, the same classification regulation applicable to the Lumina CGM System.
iCGM Special Controls
Devices classified as iCGMs must comply with specific special controls established in the De Novo classification order, including but not limited to:
• Demonstrated accuracy meeting defined performance criteria across multiple glucose ranges (hypoglycemic, euglycemic, and hyperglycemic)
• Demonstrated reliability and availability metrics
• Cybersecurity documentation meeting specified requirements
• Labeling that identifies the device as an iCGM and that includes interoperability information
• Performance testing demonstrating suitability for integration with secondary display devices and, potentially, AID systems
Lumina CGM System Performance Alignment with iCGM Criteria
The Lumina CGM System's performance targets have been designed with the iCGM classification criteria in mind. Specifically:
• The primary accuracy target of MARD ≤ 9.0% overall is consistent with the accuracy expectations described in the FDA guidance document Integrated Continuous Glucose Monitoring Systems (September 2023).
• The secondary endpoint target of ≥ 99% of paired data points in Zones A+B on the Clarke Error Grid is consistent with iCGM accuracy requirements.
• The measurement range of 40–400 mg/dL, warm-up time of ≤ 60 minutes, and alert latency of ≤ 5 minutes are designed to meet the performance expectations for iCGM-classified devices.
Strategic Considerations. Solstice is evaluating two potential regulatory strategies with respect to the iCGM classification:
Option A — Pursue iCGM Classification in the Initial 510(k) Submission: Under this approach, Solstice would submit the 510(k) with a request for iCGM classification, demonstrating compliance with all applicable special controls and iCGM performance criteria. This approach would enable the Lumina CGM System to be marketed as an iCGM from the date of clearance, permitting interoperability claims and potentially facilitating integration with AID systems.
Option B — Pursue Standard CGM 510(k) Clearance First, with Subsequent iCGM Designation: Under this approach, Solstice would initially seek 510(k) clearance for the Lumina CGM System as a standard (non-iCGM) CGM device, and would subsequently pursue iCGM classification through a supplemental or new 510(k) submission once the device has accumulated sufficient real-world performance data and interoperability testing has been completed.
Solstice seeks FDA's guidance on the preferred approach and on the implications of each pathway for the scope of the initial 510(k) submission.
FDA Reclassification Order (March 2024)
Solstice is aware that FDA issued a reclassification order effective March 2024 under 21 C.F.R. § 862.1355 that may affect the classification analysis for CGM devices, including iCGM-designated devices. Solstice seeks FDA's guidance on whether and how this reclassification order affects the iCGM classification pathway for the Lumina CGM System.
Questions