Novel SaMD, No Predicate: How to Find Your US Regulatory Pathway

When sponsors develop a novel Software as a Medical Device (SaMD) with no clear predicate, determining the correct US regulatory pathway can be a significant challenge. The initial step is a rigorous definition of the device's Intended Use and Indications for Use, which are fundamental to its classification and risk profile. If an exhaustive search of FDA databases confirms the absence of a legally marketed predicate device, the SaMD is automatically considered a high-risk Class III device. However, for novel devices that present low-to-moderate risk, the De Novo classification request offers an alternative pathway to market. This route requires a sponsor to demonstrate that general controls, or a combination of general and special controls, can provide a reasonable assurance of safety and effectiveness. This process involves a comprehensive risk analysis and robust performance data. Engaging with the FDA through the Q-Submission program is a critical best practice to gain alignment on the proposed pathway and testing strategy before committing to a final submission. ### Key Points * **Intended Use is Paramount:** The specific statements defining what the SaMD does and for whom it is intended are the most critical factors in determining its regulatory classification and risk level. * **A "Predicate" is a Specific Legal Standard:** A predicate is not just a similar device; it is a legally marketed device with the same intended use and similar technological characteristics, to which substantial equivalence can be demonstrated. A thorough search of FDA's databases is required to confirm one does not exist. * **No Predicate Means Automatic Class III:** By default, any new device type that has not been previously classified is automatically designated as Class III, requiring a Premarket Approval (PMA) application. * **De Novo is the Pathway for Novel, Lower-Risk Devices:** The De Novo pathway allows a novel, low-to-moderate risk device to be classified into Class I or II, creating a new regulatory classification for future similar devices. * **Risk Analysis is the Core of a De Novo Request:** A successful De Novo submission hinges on a robust risk analysis that identifies all potential hazards and demonstrates how they are mitigated to an acceptable level through general and/or special controls. * **Early FDA Engagement is Crucial:** Using the Q-Submission program to discuss a potential De Novo pathway with the FDA is essential. This proactive step helps de-risk the regulatory strategy and ensures alignment on classification and data requirements before significant resources are invested. ## Step 1: Defining the SaMD's Intended Use and Indications for Use The foundation of any regulatory strategy is a precise and carefully worded Intended Use statement and its associated Indications for Use. These statements dictate the device's boundaries and are scrutinized by the FDA to determine its classification. * **Intended Use:** This defines the general purpose of the device. For an SaMD, it describes what the software is intended to do (e.g., "to analyze medical images," "to monitor physiological data"). * **Indications for Use:** This statement is more specific, describing the disease or condition the SaMD will diagnose, treat, mitigate, cure, or prevent, as well as the target patient population (e.g., "...to identify signs of diabetic retinopathy in adults with diabetes"). For a novel SaMD, the level of clinical significance in these statements directly impacts risk. A SaMD intended to "inform" a clinical decision carries a different risk profile than one intended to "diagnose" a life-threatening condition. Sponsors must draft these statements with precision, as they will be the primary lens through which the FDA assesses the device. ## Step 2: Conducting an Exhaustive Predicate Search Before concluding that a device is novel, sponsors must perform a diligent search for a potential predicate device. The goal is to determine if another legally marketed device exists that could serve as a basis for a 510(k) submission. ### How to Conduct the Search 1. **FDA Product Classification Database:** Start by searching for existing product codes and classification regulations (found in 21 CFR Parts 862-892) that might apply to the SaMD's function. 2. **FDA 510(k) Premarket Notification Database:** Search for devices cleared under potentially relevant product codes. Use keywords related to the device's function, technology, and clinical application. 3. **FDA De Novo Classification Database:** Review devices that have been granted marketing authorization through the De Novo pathway. This can reveal how the FDA has classified other novel technologies. ### What to Look For A valid predicate must have the same intended use and similar technological characteristics. If the technology is different, the sponsor must demonstrate that the new technology does not raise different questions of safety and effectiveness. For a novel SaMD, particularly one using AI/ML, finding a predicate with sufficiently similar technological characteristics can be the primary obstacle. ## Step 3: Evaluating the De Novo Pathway If the predicate search confirms that no valid predicate exists, the device is automatically considered Class III. For a novel, low-to-moderate risk SaMD, the next step is to evaluate its suitability for the De Novo pathway. The central question is: **Can a reasonable assurance of safety and effectiveness be established through general and special controls?** * **General Controls:** These are the baseline requirements for all medical devices, including establishment registration, device listing, quality system regulation (21 CFR Part 820), and labeling requirements. * **Special Controls:** These are additional, device-specific controls required for Class II devices. For SaMD, special controls could include rigorous software verification and validation, performance testing standards, cybersecurity measures, and specific labeling instructions. A De Novo request is essentially a detailed argument, supported by evidence, that such controls are sufficient to mitigate the device's risks without the need for a PMA. ## Step 4: Building the De Novo Submission: Key Components A compelling De Novo request for an SaMD must be built on a foundation of robust evidence and clear justification. ### Comprehensive Risk Analysis Following a framework like ISO 14971, sponsors must identify all foreseeable risks associated with the SaMD. This includes risks from software flaws, incorrect outputs, cybersecurity vulnerabilities, and user error. For each risk, the sponsor must propose and justify specific mitigation measures that map to general or proposed special controls. ### Performance Data and Validation Unlike a 510(k) that relies on comparison to a predicate, a De Novo requires a demonstration of the device's standalone performance. For a diagnostic or analytical SaMD, this typically includes: * **Analytical Validation:** Confirms the SaMD accurately and reliably processes input data and generates technically correct output. This involves assessing the algorithm's performance using curated datasets. * **Clinical Validation:** Measures the device's performance against a clinical ground truth, demonstrating it provides clinically meaningful and accurate results for the target population and intended use. The scope and scale of this validation are critical points for discussion with the FDA. ### Proposed Special Controls The sponsor must formally propose a set of special controls that, in addition to general controls, will provide a reasonable assurance of safety and effectiveness. These proposals should be specific and actionable, becoming the basis for the new classification regulation if the De Novo is granted. ## Scenarios: Applying the Framework ### Scenario 1: A Novel AI-Based Diagnostic SaMD * **Device:** An AI algorithm that analyzes retinal images to screen for a specific, early-stage eye disease not currently screened for by software. * **Predicate Search:** Finds image management software (PACS) and viewers, but no device with the same diagnostic intended use. It is determined to be novel. * **Pathway Analysis:** The risk is moderate—a false negative could delay diagnosis, and a false positive could lead to unnecessary follow-up. The sponsor believes this risk can be mitigated with robust controls. A De Novo is the likely pathway. * **FDA Scrutiny:** The FDA would focus heavily on the clinical validation study design, the algorithm's performance (sensitivity, specificity), the diversity and quality of the training/testing datasets, and the clarity of the user labeling. ### Scenario 2: A SaMD for Patient-Driven Therapy Management * **Device:** A mobile application that uses data from a patient's glucose meter to provide personalized, non-prescriptive lifestyle and dietary coaching to help manage their condition. * **Predicate Search:** Finds apps for logging data but none that provide automated, algorithm-driven coaching for therapy management. It is determined to be novel. * **Pathway Analysis:** The risk is likely low-to-moderate. The key risk is providing poor advice that negatively impacts the patient's health. The sponsor argues this can be mitigated through software validation, human factors testing, and clear labeling that the app does not provide medical advice. A De Novo is a plausible pathway. * **FDA Scrutiny:** The focus would be on software verification and validation, human factors engineering to ensure the user interface is clear, and labeling that explicitly defines the SaMD's role as a supportive tool, not a replacement for professional medical care. ## Strategic Considerations and the Role of Q-Submission For any novel SaMD, the Q-Submission program is an indispensable strategic tool. A Pre-Submission (Pre-Sub) meeting allows a sponsor to present their proposed pathway and supporting rationale to the FDA for feedback before finalizing their submission. A Pre-Sub package for a potential De Novo device should include: * A detailed description of the SaMD, including its algorithm and principles of operation. * The precise, proposed Intended Use and Indications for Use statements. * A summary of the predicate search methodology and results. * A formal proposal for classification (Class II via De Novo) and a list of proposed special controls. * A comprehensive risk analysis summary. * An outline of the planned analytical and clinical validation studies. * Specific, well-formulated questions for the FDA regarding the appropriateness of the De Novo pathway and the sufficiency of the proposed data package. This early engagement provides invaluable feedback that can prevent costly missteps and ensure the final submission is aligned with agency expectations. ## Finding and Comparing Providers As medical device companies grow, they often require a range of specialized services to ensure compliance in different markets, from regulatory submissions to fiscal representation. Finding the right partner is crucial for navigating complex requirements efficiently. When selecting a service provider, it is important to assess their expertise, track record, and understanding of your specific product type and target markets. To find qualified vetted providers [click here](https://cruxi.ai/regulatory-directories/vat_fiscal_rep) and request quotes for free. ## Key FDA References When preparing a submission, sponsors should always consult the latest versions of official FDA documents. Key resources for novel SaMD include: * **FDA's Q-Submission Program Guidance:** Outlines the process for requesting feedback from the agency, including Pre-Submissions. * **FDA's De Novo Classification Process Guidance:** Details the requirements and procedures for submitting a De Novo request. * **21 CFR Part 807, Subpart E:** The regulation covering premarket notification (510(k)) procedures. * **FDA Guidance on Software as a Medical Device (SaMD): Clinical Evaluation:** Provides a framework for planning and executing the clinical validation of an SaMD. * **FDA Guidance on AI/ML-Based SaMD:** Discusses regulatory considerations for artificial intelligence and machine learning in medical devices. Sponsors should regularly check the FDA website for the most current guidance documents and regulations. *** This article is for general educational purposes only and is not legal, medical, or regulatory advice. For device-specific questions, sponsors should consult qualified experts and consider engaging FDA via the Q-Submission program. --- *This answer was AI-assisted and reviewed for accuracy by Lo H. Khamis.*