Does This US IVD Study Need an IDE, an IND, or Neither?

What This US IVD Study-Identity Worksheet Marks

When medical device regulatory affairs (RA) and clinical operations teams design a clinical validation protocol in the United States, they must answer one foundational question before recruiting patients or procuring human biospecimens: Does this specific study protocol qualify for an Investigational Device Exemption (IDE) exemption under 21 CFR 812.2(c)(3), proceed under abbreviated IDE requirements under 21 CFR 812.2(b), require a formal significant-risk IDE application approved by the FDA under 21 CFR 812.20, or fall under the Investigational New Drug (IND) framework of 21 CFR Part 312?

The practical answer depends on a strict, sequential decision process. You must mark product identity first—separating medical devices governed by the Federal Food, Drug, and Cosmetic Act (FD&C Act) from drugs and biological products licensed under section 351 of the Public Health Service (PHS) Act. If the investigational article is an in vitro diagnostic (IVD) device, you must evaluate the four cumulative conditions of 21 CFR 812.2(c)(3) alongside the shipment labeling mandates of 21 CFR 809.10(c). If all conditions are met, the protocol is exempt from the vast majority of Part 812 (though not from investigator disqualification under 812.119). If even one condition fails, you must assess study risk under 21 CFR 812.3(m) to select between an abbreviated IDE and a full FDA-approved IDE application.

This worksheet exists to keep several identity errors from being treated as the same mark:

• Equating leftover biobank specimens with an exemption from all human subject protections: Testing de-identified leftover blood or tissue is noninvasive under 21 CFR 812.3(k), but the donor remains a human subject under 21 CFR 812.3(p), making IRB review under 21 CFR Part 56 legally mandatory.

• Confusing Research Use Only (RUO) or Investigational Use Only (IUO) labels with regulatory exemption: Affixing an RUO or IUO statement under 21 CFR 809.10(c)(2) is shipment labeling for an investigation that is not subject to part 812. It is not proof that the protocol meets 812.2(c)(3).

• Assuming an open therapeutic IND automatically covers an investigational companion diagnostic: An investigational diagnostic used for critical treatment assignment in a pharmaceutical trial is evaluated independently under device risk criteria; an IND does not eliminate device IDE obligations.

• Treating FDA guidance as binding statute: Agency publications—such as FDA's June 2010 IVD Device Studies FAQ and April 2006 leftover-specimen guidance—are nonbinding except where they restate a regulation. Prefer current 21 CFR text as displayed on eCFR, which is unofficial, when a guidance interpretation and the regulation differ.

Device Investigation, 312 Investigation, or Listed CBER IVD Biologic

The initial determination for any clinical testing protocol is jurisdictional identity. Under 21 CFR 812.2(a), Part 812 applies to all clinical investigations of medical devices to determine safety and effectiveness, except as exempted under 812.2(c). Conversely, under 21 CFR 312.2(a), 21 CFR Part 312 governs clinical investigations of products subject to section 505 of the FD&C Act (new drugs) or biological products licensed under section 351 of the PHS Act.

If the object of the investigation is an IVD device, 21 CFR 812.2(a) applies part 812 to the clinical investigation to determine safety and effectiveness, except as provided in 812.2(c). Running the assay in a clinical laboratory, or embedding it in a multi-center therapeutic study, does not by itself move the investigation onto part 312. Listed CBER IVD biologics, combination products, and companion diagnostics remain separate identity fields.

The 21 CFR 312.2(b)(2) CBER IVD-Biologic IND Exemption

A critical regulatory carve-out often confused with device exemptions is 21 CFR 312.2(b)(2). This provision exempts clinical investigations of specific in vitro diagnostic biological products from Part 312 IND requirements. It is strictly limited to three enumerated categories regulated by the Center for Biologics Evaluation and Research (CBER):

1. Blood grouping serum;

2. Reagent red blood cells; and

3. Anti-human globulin.

To qualify for the 312.2(b)(2) IND exemption, the clinical investigation must satisfy two mandatory conditions:

• Confirmatory diagnostic use: The biological product must be intended for use in a diagnostic procedure that confirms the diagnosis made by another, medically established diagnostic product or procedure; and

• Compliant investigational shipment: The product must be shipped in full compliance with 21 CFR 312.160. Under 312.160(a)(1)(ii), the immediate package label must bear the specific text: "CAUTION: Contains a biological product for investigational in vitro diagnostic tests only."

Section 312.2(b)(3) separately exempts investigational drugs intended solely for in vitro tests or laboratory research animals, provided they are shipped under 312.160 and labeled that they are not for human use. Sponsors must recognize that 312.2(b)(2) is an IND exemption for listed biological reagents—it is not an IDE exemption under 812.2(c)(3), and it cannot be applied to molecular assays, immunoassay analyzers, or novel biomarker test kits.

Drug-Device Combination Products and Companion Diagnostics

When an IVD device is co-developed or co-evaluated with a therapeutic drug or biologic, jurisdictional assignment follows specific principles established by FDA's Office of Combination Products (OCP) and specialized cross-center guidance:

• Combination Products: FDA's combination-product FAQ states that one investigational application is generally sufficient for a combination product and should include information on the entire product. The sponsor should take into account each constituent part as well as the combination product when deciding whether an IND or IDE is needed. Typically an IND is submitted if the combination product has a drug or biologic primary mode of action and an IDE if it has a device primary mode of action. That typical assignment is not a rule that an IND covers an investigational companion diagnostic. See our dedicated analysis of combination products regulation.

• Companion Diagnostics (CDx): Unlike single-entity combination products, an in vitro companion diagnostic and its corresponding therapeutic product are distinct articles. FDA's August 2014 guidance, In Vitro Companion Diagnostic Devices, states that an investigational IVD companion diagnostic used to make critical treatment decisions—such as patient selection, treatment assignment, or treatment arm—generally will be considered a significant risk device under 21 CFR 812.3(m)(3).

The table below delineates the regulatory boundaries across product classes:

The Four 812.2(c)(3) Testing Conditions Plus 809.10(c)

When an investigation involves an IVD medical device, sponsors must evaluate 21 CFR 812.2(c)(3) to determine whether the study is exempt from most of the IDE regulation. Section 812.2(c) states that Part 812—with the exception of 21 CFR 812.119 (disqualification of clinical investigators)—does not apply to clinical investigations of diagnostic devices that meet strict testing and labeling criteria.

To secure this exemption, the sponsor must satisfy a threshold labeling predicate and four cumulative testing conditions. If the protocol fails even one condition, the exemption is void in its entirety:

1. Threshold Predicate — Compliance with 21 CFR 809.10(c): The sponsor must comply with applicable requirements in 21 CFR 809.10(c). For an investigation that 812.2(c) keeps off most of part 812, that labeling identity is 809.10(c)(2) RUO or IUO, not the 812.5 CAUTION statement used when part 812 applies.

2. Condition (i) — The Testing Is Noninvasive: Condition (i) is the 812.3(k) noninvasive definition: the diagnostic device or procedure does not, by design or intention, penetrate or pierce the skin or mucous membranes of the body, the ocular cavity, or the urethra, or enter the ear, nose, mouth, anal canal, or vagina beyond the limits 812.3(k) names. Simple venipuncture and surplus specimens left over from noninvestigational collection are noninvasive under that definition. That finding is only 812.2(c)(3)(i).

3. Condition (ii) — No Invasive Sampling Presenting Significant Risk: The testing must not require an invasive sampling procedure that presents significant risk. 812.3(k) does not skip this row. FDA's June 2010 IVD studies FAQ (Question 4) is guidance interpreting 812.2(c)(3)(ii): it recommends basing the risk determination on the harm that may result from sampling, and treats biopsy of a major organ, use of general anesthesia, or placement of a blood access line into an artery or large vein (subclavian, femoral, or iliac) as sampling that presents significant risk.

4. Condition (iii) — No Introduction of Energy: The testing must not, by design or intention, introduce energy into a subject. 812.2(c)(3)(iii) does not enumerate energy types. Operational examples such as ionizing radiation, electrical current, therapeutic ultrasound, or thermal loads are identity checks against that text, not a separate energy catalog.

5. Condition (iv) — Confirmatory Diagnostic Standard: The investigational device must not be used as a diagnostic procedure without confirmation of the diagnosis by another, medically established diagnostic product or procedure. This is the clinical management boundary.

It is vital to recognize that 21 CFR 812.119 survives the 812.2(c)(3) exemption. 812.119 investigator disqualification remains even when the rest of part 812 does not apply. FDA's IDE FAQ restates that IDE-exempt studies are not exempt from 812.119. Using a disqualified investigator is a 812.119 problem; this worksheet does not assign premarket-data rejection rates.

The table below details the four testing conditions and their operational failure modes:

The flowchart below marks product identity first, then the 812.2(c)(3) conditions, then abbreviated versus significant-risk IDE. Combination-product and companion-diagnostic rows stay in the identity table:

flowchart TD
    A["US IVD study protocol"] --> B{"Object of the investigation?"}
    B -- "Drug or biologic under 312.2(a)" --> C{"312.2(b)(2) listed CBER IVD biologic?"}
    C -- "No" --> E["Part 312 IND identity unless another 312.2(b) exemption applies"]
    C -- "Yes" --> C2{"Confirmatory use and 312.160 shipment?"}
    C2 -- "Yes" --> D["312.2(b)(2) exempt from Part 312; ship under 312.160"]
    C2 -- "No" --> E
    B -- "IVD device under 812.2(a)" --> F{"809.10(c) and all four 812.2(c)(3) conditions?"}
    F -- "Yes" --> G["812.2(c)(3) exempt; 812.119 remains; parts 50 and 56 remain"]
    F -- "No" --> H{"Significant risk under 812.3(m)?"}
    H -- "No, and not banned" --> I["812.2(b) abbreviated IDE unless FDA notifies under 812.20(a)"]
    H -- "Yes" --> J["812.20 IDE application; begin only as 812.30(a) allows"]

Leftover or Venipuncture Specimens Still Make a Human Subject

A frequent area of operational confusion in clinical laboratories centers on specimen procurement. Diagnostic manufacturers often ask whether utilizing leftover, discarded clinical specimens eliminates all human subjects research requirements.

The answer requires reading two definitions in 21 CFR 812.3 together:

• 21 CFR 812.3(k) — Noninvasive Definition: Under 812.3(k), simple venipuncture is noninvasive, and surplus body-fluid or tissue samples left over from samples taken for noninvestigational purposes are also noninvasive. That finding is only 812.2(c)(3)(i). It is not the exemption, not a skip of 812.2(c)(3)(ii)-(iv), and not a skip of 809.10(c). Condition (ii) remains a separate mark: whether the protocol requires an invasive sampling procedure that presents significant risk.

• 21 CFR 812.3(p) — Subject Definition: Section 812.3(p) defines a subject as a human who participates in an investigation, "either as an individual on whom or on whose specimen an investigational device is used or as a control."

Because 812.3(p) includes a human on whose specimen an investigational device is used, leftover-specimen testing in an IVD investigation still involves a human subject. 21 CFR 50.1 and 21 CFR 56.101 apply to clinical investigations regulated under FD&C Act section 520(g) and to investigations that support applications for research or marketing permits for medical devices. FDA's IDE FAQ states that studies exempt from the IDE regulations are not exempt from IRB review under part 56 or informed consent under part 50.

In Question 1 of the Human Subjects section of its June 2010 FAQ, FDA reaffirms this principle: an IVD study using human specimens involves human subjects. Therefore, the mere fact that specimens are leftover from routine clinical care does not excuse the sponsor from obtaining Institutional Review Board oversight under 21 CFR Part 56.

Confirmation of Diagnosis Is a Treatment-Decision Identity

Condition (iv) of 21 CFR 812.2(c)(3) represents the core clinical boundary separating exempt observational research from high-risk interventional clinical trials. The regulation mandates that the testing "is not used as a diagnostic procedure without confirmation of the diagnosis by another, medically established diagnostic product or procedure."

In clinical practice, confirmation of diagnosis is a treatment-decision identity. In its June 2010 FAQ (Question 6), FDA clarifies that investigators must rely exclusively on medically established diagnostic standards to make all clinical management and therapeutic decisions regarding human subjects. Investigational test results must never be permitted to influence patient care, therapy selection, dosing, surgical intervention, or discharge before an established, validated diagnostic establishes the condition.

The Novel Technology and Technological Advance Dilemma

A severe trap occurs when an investigational IVD incorporates a substantial technological advance over existing clinical tools. If an assay detects disease earlier, with higher analytical sensitivity, or via a novel biomarker mechanism that existing diagnostics cannot detect, medically established products may be scientifically incapable of confirming the investigational finding.

In Question 6 of the June 2010 guidance, FDA provides a concrete clinical example: an antibody test cannot confirm an investigational nucleic acid test (NAT) designed to detect early viral infection during the seronegative window period. Because the established antibody assay will yield a negative result while the investigational NAT is positive, the established test cannot confirm the NAT finding. If clinical investigators utilize the NAT result to make treatment decisions (e.g., initiating antiviral therapy or withholding blood donation), the protocol fails 812.2(c)(3)(iv).

Similarly, Question 7 addresses scenarios where no medically established diagnostic exists for a condition. If no reference method exists and investigators report the investigational result to clinicians who act upon it, the study is disqualified from the 812.2(c)(3) exemption. The sponsor must then determine whether the unconfirmed result presents potential for serious risk under 812.3(m), triggering the need for an FDA-approved IDE application under 812.20.

RUO, IUO, and 812.5 Investigational-Device Labels Are Not Interchangeable

812.2(c)(3) applies only if the sponsor complies with applicable requirements in 809.10(c). Sponsors still have to mark which labeling identity applies: 809.10(c)(2) RUO or IUO when the investigation is not subject to part 812, or 809.10(c)(1) plus 812.5 when part 812 applies.

Under 21 CFR 809.10(c), FDA establishes a bifurcated framework for in vitro diagnostic shipments:

• 21 CFR 809.10(c)(1) — Investigations Subject to Part 812: This section governs diagnostic shipments for clinical investigations that are subject to 21 CFR Part 812 (i.e., nonsignificant risk studies under 812.2(b) or significant risk studies under 812.20). Such devices must comply with Part 812 labeling, specifically bearing the investigational caution statement required by 21 CFR 812.5.

• 21 CFR 809.10(c)(2) — Investigations Not Subject to Part 812: This section applies exclusively to shipments for investigations that are exempt from Part 812 under section 812.2(c). It establishes two distinct labeling identities based on product development stage:

1. Research Use Only (RUO) under 809.10(c)(2)(i): For a product in the laboratory research phase of development that is not represented as an effective IVD, the label must bear the prominent statement: "For Research Use Only. Not for use in diagnostic procedures."

2. Investigational Use Only (IUO) under 809.10(c)(2)(ii): For a product being shipped or delivered for product testing prior to full commercial marketing (such as clinical validation studies meeting 812.2(c)(3)), the label must bear the prominent statement: "For Investigational Use Only. The performance characteristics of this product have not been established."

Contrast these exempt-study statements with 21 CFR 812.5(a), which applies when a study is governed by Part 812. Under 812.5(a), the immediate package must bear the statement:

"CAUTION—Investigational device. Limited by Federal (or United States) law to investigational use."

The 812.5 label must also disclose the manufacturer/distributor identity, relevant contraindications, hazards, adverse effects, warnings, and precautions. Furthermore, under 812.5(b), investigational labeling must never represent the device as safe or effective for the purpose under investigation.

The table below contrasts the distinct legal identities of investigational labels:

If 812.2(c)(3) Fails: Abbreviated IDE or Significant-Risk IDE Application

If a proposed clinical protocol fails any of the four conditions of 21 CFR 812.2(c)(3)—for example, if specimen collection requires an invasive organ biopsy, or if unconfirmed results will guide surgical margins or systemic therapy—the investigation is subject to 21 CFR Part 812. The sponsor must immediately conduct a risk assessment under 21 CFR 812.3(m) to determine whether the study represents a Nonsignificant Risk (NSR) or Significant Risk (SR) investigation.

Evaluating Significant Risk under 21 CFR 812.3(m)

Under section 812.3(m), a significant risk device is defined as an investigational device that:

1. Is intended as an implant and presents a potential for serious risk to the health, safety, or welfare of a subject;

2. Is purported or represented to be for a use in supporting or sustaining human life and presents such a potential;

3. Is for a use of substantial importance in diagnosing, curing, mitigating, or treating disease, or otherwise preventing impairment of human health, and presents such a potential; or

4. Otherwise presents a potential for serious risk to the health, safety, or welfare of a subject.

The June 2010 IVD studies FAQ is guidance interpreting 812.3(m) for IVDs. It interprets potential for serious risk in relation to the harm that may result from misdiagnosis or error in treatment, including life-threatening harm or permanent impairment, and discusses false-positive and false-negative consequences:

• False-Positive Harms: The June 2010 FAQ states that false-positive results can lead to unnecessary confirmatory testing, unnecessary treatment that can be invasive or have harmful side effects, and/or unnecessary psychological trauma when serious or life-threatening diseases or conditions are involved.

• False-Negative Harms: The same FAQ treats false-negative consequences as delayed or missed diagnosis and error in treatment. Those examples are guidance, not numeric harm thresholds.

Abbreviated IDE Requirements under 21 CFR 812.2(b)

If 812.2(c)(3) does not apply and the investigation is of a device other than a significant risk device, 21 CFR 812.2(b) treats the investigation as having an approved IDE if the device is not a banned device, FDA has not notified the sponsor under 812.20(a) that an application is required, and the sponsor complies with the abbreviated duties:

• Device Labeling: The device must be labeled in full accordance with 21 CFR 812.5;

• IRB Approval: The sponsor must obtain IRB approval after presenting the reviewing IRB with a brief explanation of why the device is not a significant risk device, and maintain that approval;

• Informed Consent: The sponsor must ensure that each participating investigator obtains informed consent from each subject under the investigator's care in accordance with part 50;

• Study Monitoring: The sponsor must monitor the trial in compliance with 21 CFR 812.46;

• Records and Reports: The sponsor must maintain the records required under 812.140(b)(4) and (5) and make the reports required under 812.150(b)(1) through (3) and (5) through (10);

• Investigator records and reports: The sponsor must ensure that participating investigators maintain the records required by 812.140(a)(3)(i) and make the reports required under 812.150(a)(1), (2), (5), and (7); and

• Prohibition on Commercialization: The sponsor must comply with the prohibitions in 812.7 against promotion and other practices.

Significant Risk IDE Applications under 21 CFR 812.20

If the investigation involves a significant risk device, 21 CFR 812.20 mandates that the sponsor submit a formal IDE application to FDA. Under 812.20(a)(2), the sponsor shall not begin the investigation until FDA has approved the application. 812.20(a)(1) also requires an application if the investigation involves an exception from informed consent under 21 CFR 50.24, or if FDA notifies the sponsor that an application is required. Under 812.30(a), an investigation may not begin until 30 days after FDA receives the application unless FDA notifies the sponsor that the investigation may not begin, or until FDA approves the IDE by order. 812.30(a) allows FDA to approve the investigation as proposed, approve it with modifications, or disapprove it. This article does not assign typical review clocks or approval rates.

Crucially, under 21 CFR 812.66, if an IRB determines that an investigation presented for approval under 812.2(b)(1)(ii) involves a significant risk device, it shall notify the investigator and, where appropriate, the sponsor. A sponsor may not begin the investigation except as provided in 812.30(a). An IRB nonsignificant-risk concurrence is not a 812.2(c)(3) exemption.

Leftover-Specimen Informed-Consent Discretion Is Not an IRB Skip

One of the most consequential compliance breakdowns in IVD premarket programs involves misinterpreting FDA's April 2006 guidance, Guidance on Informed Consent for In Vitro Diagnostic Device Studies Using Leftover Human Specimens that are Not Individually Identifiable. Many sponsors mistakenly view this guidance as an exemption from all regulatory oversight. In reality, it is an enforcement discretion policy strictly addressing informed consent under 21 CFR Part 50; it provides zero exemption from Institutional Review Board review under 21 CFR Part 56.

When 2006 leftover-specimen informed-consent discretion is even in scope

The April 2006 leftover-specimen guidance is enforcement discretion for informed consent, not an IDE exemption and not an IRB skip. Under Section IV, FDA intends to exercise that discretion if all of the following Section IV conditions are true:

1. 812.2(c)(3) already met: The investigation meets the IDE exemption criteria at 21 CFR 812.2(c)(3).

2. Leftover remnants, or repository / unrelated-research leftovers: The study uses leftover specimens—remnants collected for routine clinical care or analysis that would have been discarded. The policy also includes specimens obtained from specimen repositories and leftovers previously collected for other research purposes.

3. Not individually identifiable: The identity of the subject is not known to, and may not readily be ascertained by, the investigator or any other individuals associated with the investigation, including the sponsor. If the specimen is coded, the 2006 guidance treats it as not individually identifiable only if neither the investigators nor any other individuals associated with the investigation or the sponsor can link the specimen to the subject, either directly or indirectly through coding systems.

4. Clinical information does not identify the source: Specimens may be accompanied by clinical information only if that information does not make the specimen source identifiable to the investigator or any other individual associated with the investigation, including the sponsor.

5. Caregivers are separated from the investigation: The individuals caring for the patients are different from, and do not share information about the patient with, those conducting the investigation.

6. Supplier identifier controls: The specimens are provided to the investigator without identifiers, and the supplier has established policies and procedures to prevent the release of personal information.

7. IRB review remains: The study has been reviewed by an IRB in accordance with 21 CFR Part 56, except as described in section 7 of the 2006 guidance, which addresses IRB informed-consent duties under the same discretion policy and does not skip part 56.

The same 2006 section says studies outside this discretion include, but are not limited to, investigations that do not meet 812.2(c)(3); specimens that are individually identifiable; specimens collected specifically for the proposed investigation; more specimen than would leftover from usual clinical collection; or test results reported to the subject's health care provider. When discretion is out of scope, informed consent under 21 CFR Part 50 applies unless a listed part 50 exception independently applies. Waiver and exception mechanics are not this worksheet's identity decision. For consent mechanics, see our dedicated guide to informed consent in medical device clinical trials.

The FDA 18 October 2021 Letter to Industry

On 18 October 2021, FDA's Center for Devices and Radiological Health issued a formal Letter to Industry: Studies Using Leftover, Deidentified Human Specimens Require IRB Review. The letter reminds IVD manufacturers that FDA requires IRB review for all clinical investigations of devices that involve human subjects, including leftover, deidentified human specimens used in FDA-regulated studies, and that the 2006 leftover-specimen guidance does not exempt any investigation from IRB requirements.

The 2021 letter restates these boundaries:

• IRB review remains required: The 2021 letter states that IRB review is required for all clinical investigations of devices that involve human subjects, including leftover, deidentified human specimens in FDA-regulated studies, and that this requirement pertains to data used to support an IDE, device marketing application, or submission, including IVD technical or analytical studies that use human specimens.

• Consent Discretion Does Not Equal IRB Exemption: The April 2006 guidance addresses only the enforcement of informed consent requirements under 21 CFR Part 50. The guidance does not, and legally cannot, exempt sponsors from IRB requirements under 21 CFR Part 56.

• Expedited Review Remains an IRB Action: Whether an IRB uses expedited procedures under 21 CFR 56.110 is an IRB action. A sponsor cannot self-certify expedited review or treat leftover specimens as skipping part 56.

Labeled Boundaries This Worksheet Does Not Decide

The following provisions are labeled boundaries. They are not this protocol-identity worksheet:

• Marketed Indications under 21 CFR 812.2(c)(1)-(2): 812.2(c)(1) and (c)(2) exempt investigations of certain devices in commercial distribution when used according to the labeled indications those provisions name. They are not the 812.2(c)(3) diagnostic exemption.

• Consumer Preference and Customer Modifications under 21 CFR 812.2(c)(4): Testing of consumer preference, modifications to existing devices, or combinations of marketed devices are exempt under 812.2(c)(4) only if the testing is not for the purpose of determining safety or effectiveness and does not put subjects at risk.

• Veterinary Devices under 21 CFR 812.2(c)(5): Devices intended solely for veterinary diagnostic use are exempt from Part 812. They fall under veterinary device regulatory authorities and must not be evaluated under human clinical criteria.

• Transitional Provision under 21 CFR 812.2(e): 812.2(e) is a 1980 transitional IND-to-IDE provision for a sponsor that, on 16 July 1980, had an effective IND for an investigation of a device. It is not a current IND path for an IVD protocol.

• European Union IVDR Chapter VI: EU IVDR Article 58 applications, Article 70 PMPF notifications, and Article 5(5) in-house manufacture are Regulation (EU) 2017/746 decisions. They are not 812.2(c)(3), abbreviated IDE, or 312 identity.

• CLIA Verification under 42 CFR 493.1253: Clinical laboratory analytical performance verification of an unmodified FDA-cleared/approved test prior to reporting patient results is an operational CLIA quality requirement, not an investigational IDE protocol under 21 CFR Part 812.

• Draft Guidance on Investigational IVDs in Therapeutic Trials (December 2017): FDA's December 2017 draft guidance, Investigational IVDs Used in Clinical Investigations of Therapeutic Products, remains draft and not for implementation. Do not treat it as current 812.2.

• Pre-RFD center assignment: A Pre-Request for Designation asks FDA which center will regulate a product. That is not 812.2(c)(3), abbreviated IDE, significant-risk IDE, or 312 identity. See the Pre-RFD playbook.

• CLIA 42 CFR 493.1253 and LDT identity: Verification of a marketed unmodified test, and LDT-rule identity, are not this leftover-sample IDE-versus-IND worksheet.

This page does not recast the published IVDR performance-study encyclopedia, the device-only IDE guide, the cross-jurisdiction investigational-device comparison, companion-diagnostic marketing pathways, RUO/IUO labeling, the RUO-to-IVD conversion firewall, the LDT regulatory guide, the IVD encyclopedia, dual 510(k)/CLIA-waiver submissions, combination-product OCP/RFD identity, the Pre-RFD playbook, informed-consent mechanics, expanded access, or DHT-as-investigational-device identity.

Mark One Labeled Fictional Protocol Packet

The identity fields are marked on one labeled fictional protocol packet: PROTO-FICT-2026-09.

Protocol Summary: PROTO-FICT-2026-09

• Protocol Title: Clinical Evaluation of the Next-Generation Digital Droplet PCR ctDNA Liquid Biopsy Assay for Early Progression Monitoring in Non-Small Cell Lung Cancer (NSCLC).

• Investigational Device: Quant-Lung ctDNA Molecular Assay (an automated droplet digital PCR platform detecting KRAS and EGFR circulating tumor DNA variants in human plasma).

• Study Architecture (Two Comparative Arms):

1. Arm 1 (Observational Remnant Cohort): 200 banked, de-identified remnant plasma samples collected from NSCLC patients during routine standard-of-care follow-up. All assay testing is performed retrospectively. Assay results are completely blinded to treating oncologists and patients. All disease progression and treatment decisions are determined exclusively by standard computed tomography (CT) scans using RECIST 1.1 criteria.

2. Arm 2 (Prospective Interventional Cohort): 50 newly enrolled NSCLC patients undergoing systemic targeted therapy. Blood is collected specifically for the study. If the investigational ctDNA assay detects a >25% rise in mutant allele fraction, treating oncologists immediately switch patients to second-line chemotherapy or immunotherapy prior to radiographic progression on CT imaging.

Step-by-Step Worksheet Execution

Applying the identity worksheet across both arms shows how a protocol change moves the investigation from 812.2(c)(3) onto a significant-risk IDE application:

• Step 1: Product Identity: Both arms investigate an IVD device under 21 CFR 812.2(a). Neither arm is a 21 CFR 312 investigation of a drug or biologic, and the assay is not a 312.2(b)(2) listed CBER IVD biologic. The governing framework is 21 CFR Part 812.

• Step 2: 812.2(c)(3) Testing Conditions for Arm 1: Arm 1 satisfies all four conditions: (i) testing is noninvasive (performed on plasma); (ii) specimen collection was routine venipuncture involving no invasive significant-risk sampling under 812.3(k); (iii) no energy is introduced into subjects; and (iv) results are blinded and never used for clinical management, with disease status confirmed independently by established CT imaging under 812.2(c)(3)(iv). Determination: Arm 1 is 21 CFR 812.2(c)(3) EXEMPT.

• Step 3: 812.2(c)(3) Testing Conditions for Arm 2: Arm 2 satisfies conditions (i), (ii), and (iii). However, Arm 2 fails Condition (iv). The investigational ctDNA result directly alters clinical management (switching systemic therapy) without confirmation by established RECIST 1.1 radiographic criteria. Under June 2010 FAQ Q6/Q7, unconfirmed clinical management strips the exemption. Determination: Arm 2 is NON-EXEMPT and subject to Part 812.

• Step 4: Risk Assessment for Arm 2: Under 21 CFR 812.3(m)(3), altering systemic oncology therapy based on an unconfirmed diagnostic presents potential for serious risk (toxic side effects of second-line therapy or premature discontinuation of effective first-line treatment). Therefore, Arm 2 is a Significant Risk Device investigation. Under 21 CFR 812.20, the sponsor must submit a full IDE application to FDA and obtain written approval prior to enrolling patients.

• Step 5: Labeling Identity: Kits shipped for Arm 1 must bear the 21 CFR 809.10(c)(2)(ii) IUO statement: "For Investigational Use Only. The performance characteristics of this product have not been established." Kits shipped for Arm 2 must bear the 21 CFR 812.5(a) CAUTION statement: "CAUTION—Investigational device. Limited by Federal (or United States) law to investigational use."

• Step 6: Human Subject and Ethics Oversight: Under 21 CFR 812.3(p), both arms involve human subjects. Institutional Review Board review under 21 CFR Part 56 is mandatory for both Arm 1 and Arm 2. For Arm 1, the sponsor may utilize FDA's 2006 enforcement discretion for informed consent under Part 50, provided samples are genuinely de-identified. For Arm 2, prospective written informed consent under 21 CFR Part 50 is strictly required.

The comprehensive classification matrix below encapsulates the evaluation of PROTO-FICT-2026-09:

Mark the product first, then the four 812.2(c)(3) testing conditions plus 809.10(c) labeling, then leftover-specimen human-subject duties, then abbreviated versus significant-risk IDE only if the exemption fails. This worksheet is manufacturer-facing identity, not individual legal, IRB, or clinical advice.