FDA Proposes to Down‑Classify Key Oncology Companion Diagnostics to Class II: What CDx Sponsors Need to Know

Introduction: A Structural Shift in FDA Companion Diagnostic Policy

FDA has published a proposed order in the Federal Register to reclassify a major family of oncology companion diagnostics and related molecular tests—from post‑amendment Class III (PMA) devices to Class II with special controls, under a new regulation at 21 CFR 866.6075.

This proposal creates, for the first time, a unified Class II device type for “Nucleic Acid‑Based Test Systems for Use with a Corresponding Approved Oncology Therapeutic Product”. It would move many oncology nucleic‑acid tests (including core companion diagnostics) onto a 510(k) pathway—while maintaining stringent analytical and clinical expectations via special controls.

For CDx sponsors, pharma partners and oncology laboratories, this is more than a technical reclassification. It reshapes how you plan evidence, sequence FDA and EMA/IVDR strategies, budget user fees, and negotiate drug–test labeling.

As an IVD‑focused CRO with deep experience in IVDR performance studies, EMA CDx consultation and FDA submissions, MDx CRO is already working with sponsors to future‑proof pipelines against this new framework.

What Exactly Has FDA Proposed?

FDA’s proposed order would:

  • Create a new device type at 21 CFR 866.6075 titled “Nucleic Acid‑Based Test Systems for Use with a Corresponding Approved Oncology Therapeutic Product.”
  • Reclassify post‑amendment Class III devices in this category—currently approved under product codes OWD, PJG, PQP and SFL—into Class II (special controls), subject to 510(k).
  • Apply the new regulation to both currently approved devices and future devices that fit this definition.

Under the proposed identification, these devices are prescription IVDs that:

  • Detect specific genetic variants and/or other nucleic acid biomarkers in human clinical specimens
  • Use nucleic acid amplification (e.g., PCR) and/or sequencing technologies (e.g., NGS)
  • Provide information related to the use of a corresponding FDA‑approved oncology therapeutic product

Crucially, the device type covers two sub‑families:

1 – True companion diagnostics – tests essential for the safe and effective use of an oncology therapeutic product, with reciprocal references in both the device labeling and the drug labeling.

2 – “CDx‑adjacent” tests – tests that are not essential for use of the drug, but are referenced in the drug’s labeling because they provide information about known benefits and/or risks (for example, prognostic or response‑enrichment information).

    FDA has opened a 60‑day public comment period following publication of the proposed order; if finalized, the new classification would take effect 30 days after the final order is published in the Federal Register.

    Why Is FDA Comfortable Moving Oncology NAAT/NGS CDx to Class II?

    FDA’s justification rests on a decade‑plus of real‑world experience with these technologies, framed through the statutory reclassification provisions in section 513(f)(3) of the FD&C Act.

    1. Extensive PMA Experience

    Since the first approval in 2011 (cobas 4800 BRAF V600 Mutation Test), FDA has approved 35 PMAs and 403 PMA supplements for oncology nucleic acid‑based test systems across four product codes (OWD, PJG, PQP, SFL).

    For this proposed reclassification, FDA relied on data from 17 original PMAs and one panel‑track supplement, making use of the “six‑year rule” in section 520(h)(4) to draw on PMA datasets once they were sufficiently aged to be used for reclassification.

    These submissions include PCR hotspot assays, NGS panels, BRCA and RAS panels, MSI signatures and other archetypal oncology CDx platforms.

    2. Post‑market Safety and Recall Data

    FDA analysed MDRs (MAUDE database) and recalls for these product codes:

    • 147 MDRs across four product codes, with >80% related to false positives; >95% reported no clinical harm, and the fraction associated with serious health impact was low.
    • Four Class III recalls and 23 Class II recalls, with root causes concentrated in non‑specific molecular interactions, non‑conforming components, process deviations, labelling errors and off‑label use—not systemic failures of the technology itself.

    The conclusion: risk modes are well understood and can be mitigated through defined special controls, rather than case‑by‑case PMA‑level scrutiny.

    3. Technological Maturity

    FDA explicitly characterises oncology nucleic acid‑based test systems as a well‑bounded technological family, with:

    • Similar purposes and intended uses
    • Common analytical principles (PCR and NGS)
    • Comparable bioinformatics pipelines
    • No unique, unmitigated risk patterns across product codes

    On that basis, FDA finds that Class II with appropriate special controls can provide reasonable assurance of safety and effectiveness.

    What Are the Proposed Special Controls?

    The special controls in proposed 21 CFR 866.6075(b) fall into two main buckets: design verification/validation and labeling.

    1. Design Verification and Validation

    Sponsors must provide:

    Analytical performance data

    • Precision, analytical accuracy, analytical sensitivity, analytical specificity
    • Sample and reagent stability
    • Evaluation for each gene/variant, or an FDA‑accepted representative strategy

    Coverage and limitations

    • Data showing which genomic regions are targeted and detected
    • Disclosure of regions not covered or with limited detection

    Clinical performance data

    • Generated using clinical specimens representative of the intended use population
    • Demonstrating appropriate clinical performance for the intended use (e.g., response‑predictive or risk‑informative claims)

    Specimen handling validation

    • Validation of specimen types, pre‑analytical handling, extraction and preparation as described in the labeling

    Biomarker classification / bioinformatics validation

    • Clear description of the classification pipeline and criteria
    • Documentation of the basis for biomarker interpretation, with appropriate references

    Risk‑mitigation specifications

    • Controls, procedures and user‑training requirements designed to reduce false results and misinterpretation

    In practice, this codifies what high‑quality oncology CDx sponsors already do under PMA—now packaged as device‑type‑level expectations rather than bespoke PMA negotiations.

    2. Labeling Controls – Including the New Label‑Alignment Requirement

    Labeling must include:

    • Device description (biomarkers detected, analysis algorithms, quality metrics, detection limitations)
    • A summary of analytical and clinical performance studies and their results

    For essential CDx tests:

    • Language stating that the test is indicated for use with a specific FDA‑approved oncology therapeutic product, and
    • A requirement that device labeling be consistent with the corresponding drug’s FDA‑approved labeling.

    For non‑essential but drug‑informative tests:

    • Language summarising known benefits and/or risks of the drug as informed by the test, which also must be consistent with the therapeutic product’s FDA‑approved labeling.

    That last element—the requirement that test claims about drug benefit or risk “track” the drug label—is where the most interesting policy tension sits.

    Why This Matters Operationally: PMA vs 510(k) for Oncology CDx

    If finalized, reclassification would move this family of tests from PMA to 510(k), with important practical effects:

    • Less burdensome submission type. FDA itself notes that the 510(k) pathway is generally less burdensome and more cost‑effective than PMA, with shorter review timelines.
    • Lower user fees. In FY 2025, the standard user fee for a PMA‑type application is about $541k, while a 510(k) fee is about $24k—roughly a 20‑to‑1 difference, before small‑business reductions.
    • More predictable review questions. Instead of re‑litigating the entire benefit–risk story, sponsors will argue substantial equivalence and conformity to codified special controls.
    • More flexible change control. FDA explicitly points to predetermined change control plans (PCCPs) as a way to implement certain future modifications without a new 510(k) each time, if the change is covered by a cleared PCCP.

    For oncology CDx sponsors, that means:

    • Faster and cheaper routes for follow‑on tests targeting established biomarkers (EGFR, BRAF, BRCA, MSI, etc.).
    • A clearer template for analytical and clinical evidence packages.
    • Easier lifecycle management for panel expansions, algorithm updates and bioinformatics refinements—subject to PCCP design.

    The Built‑In Ceiling: Test Claims Cannot Outrun Drug Labeling

    While the new Class II device type creates a more efficient pathway, it also codifies a strict rule about how far test labeling can go:

    For tests essential to a drug’s safe and effective use, device labeling must be consistent with the corresponding drug label.

    For tests that merely provide information about known benefits or risks (without being essential), the device can only summarise those benefits/risks in language that is, again, consistent with the drug label.

    In practice, this means:

    • You cannot obtain CDx labeling for a novel biomarker–drug relationship unless the drug’s own labeling acknowledges that relationship.
    • Independent evidence generated by an IVD sponsor, academic group or cooperative trial cannot, on its own, support drug‑linked claims in device labeling if the corresponding therapeutic product label remains silent or disagrees.

    This alignment rule is not new—it has long applied in the PMA CDx world—but the new regulation explicitly extends it to the broader family of “CDx‑adjacent” oncology NAAT/NGS tests.

    From a sponsor’s perspective, that creates a structural stalemate scenario:

    • A test developer may generate high‑quality evidence showing that a biomarker predicts lack of benefit or heightened risk for a widely used oncology drug.
    • If the drug sponsor declines to pursue a label update, FDA cannot allow device labeling that effectively rewrites the drug’s benefit–risk profile.
    • The test might still be cleared as a purely analytical device, but it would not carry the clinically meaningful, drug‑linked claims that drive adoption and reimbursement.

    For CDx developers, the implication is clear:

    Regulatory strategy for these new Class II tests still hinges on drug‑sponsor alignment, not just assay science.

    What Should Sponsors Do Now?

    Even before the order is final, CDx and oncology test sponsors can act:

    1. Map your pipeline against the new device type: Identify NAAT/NGS oncology tests tied to approved therapies that fall into the new 866.6075 definition.
    2. Re‑think U.S. regulatory roadmaps: For qualifying tests, plan around a 510(k) with special controls instead of a PMA—while recognising that evidence expectations remain high.
    3. Align CDx and drug strategies early: For new biomarkers, ensure drug‑side teams understand that therapeutically meaningful test claims will require corresponding drug‑label changes.
    4. Design performance studies that satisfy both FDA and IVDR: Build analytical and clinical validation packages that can support 510(k) clearance and the IVDR Class C CDx pathway, including EMA consultation.
    5. Leverage PCCPs thoughtfully: Consider where a PCCP can safely cover bioinformatics updates, new variants or refined cut‑offs—without triggering repeated submissions.

    How MDx CRO Can Help

    MDx CRO is built around IVD and CDx:

    • We deliver end‑to‑end IVD trial solutions, including delegated sponsor responsibilities, ISO 20916‑compliant clinical performance studies, biostatistics, data management, safety reporting and IVDR‑compliant study reports.
    • Our regulatory affairs team designs global strategies for IVDR and FDA, including device classification, performance evaluation, technical documentation and regulatory submissions.
    • We are already supporting IVDR CDx EMA consultation processes, making us well placed to harmonise evidence plans with FDA’s new Class II framework for oncology NAAT/NGS CDx.

    If your portfolio includes oncology companion diagnostics, predictive biomarkers or NGS oncology panels, this proposed reclassification is an opportunity to:

    • Reduce time and cost to achieve an FDA label
    • Align U.S. and EU strategies more intelligently
    • Design performance studies that work once and serve both regimes

    Contact MDx CRO to discuss how to stress‑test your CDx roadmap against the new 21 CFR 866.6075 framework and build a regulatory plan that works on both sides of the Atlantic.

    This article is provided for general information only and does not constitute legal or regulatory advice. Sponsors should consult their own regulatory and legal counsel for device‑specific strategies.

    Written by:
    Carlos Galamba

    Carlos Galamba

    CEO

    Senior regulatory leader and former BSI IVDR reviewer with deep experience in CE marking high-risk IVDs, companion diagnostics, and IVDR implementation.
    Industry Insights & Regulatory Updates