Companion Diagnostics (CDx): Regulatory Pathways, Development, and Market Access

What Are Companion Diagnostics?

Companion diagnostics (CDx) are in vitro diagnostic devices that provide information essential for the safe and effective use of a corresponding therapeutic product. They are not optional screening tools or general-purpose laboratory tests. A companion diagnostic is, by regulatory definition, required -- the therapeutic product's labeling specifies that the CDx must be used to make a treatment decision, whether that decision is about selecting patients who will benefit, identifying patients who are likely to experience serious adverse effects, or monitoring response to adjust dosing.

The FDA defines a companion diagnostic under 21 CFR 809.3 as an in vitro diagnostic device that provides information that is essential for the safe and effective use of a corresponding therapeutic product. The word "essential" is the operative term. It means the therapeutic cannot be used safely or effectively without the diagnostic information. This distinguishes a companion diagnostic from every other category of diagnostic test.

CDx in the Context of Precision Medicine

Companion diagnostics are the operational backbone of precision medicine. Without them, the concept of "the right drug for the right patient at the right time" is an aspiration without a mechanism. Every targeted therapy requires a way to identify the target. Every biomarker-driven treatment decision requires a validated test to measure the biomarker.

The modern era of companion diagnostics began with the approval of trastuzumab (Herceptin) for HER2-positive breast cancer in 1998, paired with the HercepTest immunohistochemistry assay. That pairing established the principle: if you develop a drug that works only in a biomarker-defined subpopulation, you must simultaneously develop and obtain approval for a test that identifies that subpopulation.

Since then, the CDx landscape has expanded dramatically. As of 2026, there are more than 50 FDA-approved companion diagnostics spanning oncology, infectious disease, cardiology, and rare diseases. The majority are in oncology, where targeted therapies and immuno-oncology agents have made biomarker testing a standard-of-care requirement.

CDx vs. Complementary Diagnostics

The distinction between companion diagnostics and complementary diagnostics is not semantic -- it has direct regulatory and labeling consequences. A companion diagnostic is essential: the drug label mandates its use. A complementary diagnostic provides information that aids treatment decisions but is not required. The drug can still be prescribed without the complementary diagnostic result.

The classic example involves PD-L1 testing for checkpoint inhibitors. For pembrolizumab (Keytruda) in certain indications, PD-L1 expression testing with a specific CDx (the PD-L1 IHC 22C3 pharmDx assay) is required -- pembrolizumab is only indicated for patients whose tumors express PD-L1 above a defined threshold. In contrast, for nivolumab (Opdivo), PD-L1 testing using the PD-L1 IHC 28-8 pharmDx assay was approved as a complementary diagnostic -- it provides useful prognostic and predictive information, but nivolumab can be prescribed regardless of PD-L1 status.

CDx vs. Laboratory-Developed Tests (LDTs)

Laboratory-developed tests are diagnostic tests designed, manufactured, and used within a single laboratory. Historically, LDTs operated under the practice of medicine and were largely exempt from FDA oversight, though they were subject to regulation under the Clinical Laboratory Improvement Amendments (CLIA).

In 2024, the FDA finalized a rule phasing out its general enforcement discretion policy for LDTs, establishing a timeline for LDTs to come under full device regulation. This rule has significant implications for companion diagnostics, because many biomarker tests used in clinical treatment decisions -- including some that function as de facto companion diagnostics -- have been performed as LDTs without FDA clearance or approval.

The regulatory distinction matters enormously. An FDA-approved CDx has undergone rigorous analytical and clinical validation, has been reviewed by FDA, and has performance characteristics documented in its labeling. An LDT performing the same biomarker assay may or may not have equivalent performance, and its validation is self-certified by the laboratory.

Practical Note: When a therapeutic product's label specifies that a companion diagnostic must be used, the FDA expects the use of an FDA-approved CDx. Using an LDT in place of the specified CDx raises significant regulatory and liability questions, even if the LDT targets the same biomarker. Physicians may still exercise clinical judgment, but payers, hospitals, and pharmaceutical partners increasingly require FDA-approved CDx results.

The Regulatory Landscape

The regulatory framework for companion diagnostics has evolved substantially since the late 1990s. Today it reflects a mature -- if still complex -- system in both the United States and the European Union.

FDA Framework

The FDA's regulatory approach to CDx is built on several key documents and authorities:

• Guidance for In Vitro Companion Diagnostic Devices (2014, updated): This is the foundational guidance document. It establishes that when the safe and effective use of a therapeutic product depends on an IVD, the FDA generally will not approve or clear the therapeutic without simultaneously approving or clearing the IVD.
• 21 CFR 809.3: Provides the regulatory definition of a companion diagnostic device.
• 21 CFR 809.10: Labeling requirements for in vitro diagnostic products, including CDx-specific labeling elements.
• CDER-CDRH Parallel Review: A formal process enabling the Center for Drug Evaluation and Research (CDER) or Center for Biologics Evaluation and Research (CBER) and the Center for Devices and Radiological Health (CDRH) to coordinate their reviews of therapeutics and their companion diagnostics.

The FDA's position is clear: a companion diagnostic should be approved or cleared contemporaneously with the therapeutic product it accompanies. In practice, this means the CDx submission and the therapeutic submission are reviewed in parallel, with cross-referencing between CDRH (which reviews the CDx) and CDER or CBER (which reviews the drug or biologic).

Standalone vs. Co-Developed CDx

There are two development models for companion diagnostics:

Co-developed CDx: The diagnostic and the therapeutic are developed together from early clinical development. The CDx is used in the pivotal clinical trials for the therapeutic, and both are submitted for approval concurrently. This is the model the FDA prefers and the one most consistent with the regulatory framework.

Standalone or post-hoc CDx: The therapeutic is already approved, and a CDx is developed afterward to identify patients who would benefit. This pathway is less common and more challenging, because the CDx developer must generate clinical evidence linking the test result to therapeutic benefit, often through retrospective analyses of banked specimens from completed clinical trials.

Development Model	Advantages	Challenges
Co-developed CDx	Regulatory alignment; CDx used in pivotal trials; simultaneous approval	Requires early alignment between drug and diagnostics companies; timeline dependency
Standalone/Post-hoc CDx	Independent development timeline; can leverage existing clinical data	Must generate clinical evidence retrospectively; regulatory pathway less defined; may face specimen availability issues

FDA CDx Regulatory Pathways

Companion diagnostics are regulated as medical devices by CDRH. The specific regulatory pathway depends on the device classification, the risk level, and whether a predicate device exists.

PMA Pathway (Premarket Approval)

The premarket approval pathway is the most common for companion diagnostics. Most CDx are classified as Class III devices because they are essential for treatment decisions involving high-risk therapeutic products. A CDx that determines whether a patient receives a cancer drug is, by definition, a device whose failure could lead to serious harm -- either through inappropriate treatment of a patient who would not benefit (false positive) or denial of effective treatment to a patient who would benefit (false negative).

PMA requires the highest level of evidence: valid scientific evidence demonstrating that the device is safe and effective for its intended use. For a CDx, this typically means:

• Complete analytical validation (sensitivity, specificity, accuracy, precision, reproducibility, limit of detection, interference testing)
• Clinical validation demonstrating that the test result is associated with clinical outcomes relevant to the therapeutic product
• Clinical specimen data, often from the pivotal clinical trial of the therapeutic
• Manufacturing and quality system documentation
• Labeling that cross-references the corresponding therapeutic product

The PMA review process typically takes 180 days from FDA acceptance, though total elapsed time is often longer when additional information is requested.

Key Insight: The PMA pathway carries substantial costs -- often $500,000 to $2 million or more in user fees alone, plus the cost of clinical data generation. However, it also provides the strongest market exclusivity and the highest level of regulatory credibility, which is particularly important when payers and pharmaceutical partners are making coverage and co-marketing decisions.

510(k) Pathway

The 510(k) pathway is available for CDx that can demonstrate substantial equivalence to an existing legally marketed predicate device. This pathway is less common for companion diagnostics because most CDx address novel biomarkers or novel therapeutic indications for which no predicate exists.

However, 510(k) may be appropriate in specific scenarios:

• A new CDx that tests for the same biomarker using the same technology as an existing cleared CDx, but with minor modifications
• A CDx for a biomarker that has an established predicate in a different therapeutic context, where the analytical performance requirements are equivalent
• Certain lower-risk CDx applications where the treatment decision is not the sole determinant of patient management

The 510(k) pathway is faster (90-day review timeline) and less expensive, but it provides less regulatory exclusivity and may be viewed as less robust by pharmaceutical partners.

De Novo Pathway

The De Novo classification pathway is for novel devices that are low to moderate risk but have no predicate. For CDx, the De Novo pathway is relatively uncommon because most CDx are classified as Class III due to the critical nature of the treatment decisions they inform. However, some CDx with lower-risk profiles -- for instance, those used to inform dosing adjustments rather than binary treatment/no-treatment decisions -- may qualify.

The De Novo pathway requires a risk-based classification determination by FDA. If granted, the device receives a new classification regulation and becomes a predicate for future 510(k) submissions.

Breakthrough Device Designation

The FDA's Breakthrough Device Designation program is particularly relevant for companion diagnostics that address unmet medical needs. CDx that accompany breakthrough therapies, or that enable identification of patients for novel therapeutic modalities (such as cell and gene therapies), are strong candidates.

Breakthrough designation provides:

• More intensive FDA interaction and guidance during development
• Priority review
• Potential for rolling submission
• Senior management involvement in the review

As of 2026, several CDx have received breakthrough designation, particularly NGS-based panels and liquid biopsy assays that enable real-time treatment selection across multiple tumor types.

Summary of FDA CDx Pathways

Pathway	Class	Evidence Standard	Typical Review Timeline	User Fee (Approx.)	Common CDx Use Cases
PMA	III	Valid scientific evidence (safety and effectiveness)	180 days	$500K-$2M+	Most CDx (oncology, rare disease)
510(k)	II	Substantial equivalence	90 days	$20K-$25K	Modifications to existing CDx
De Novo	I/II	Reasonable assurance of safety and effectiveness	150 days	$120K-$150K	Novel low/moderate-risk CDx
HDE	III	Probable benefit outweighs risk	75 days	Exempt	CDx for rare diseases (<8,000 patients/year)

Proposed Down-Classification of Oncology CDx (2025)

In November 2025, the FDA published a proposed order in the Federal Register to reclassify nucleic acid-based oncology companion diagnostics from Class III (PMA) to Class II (510(k) with special controls). This is the first broad structural change to CDx oversight in more than a decade and reflects the FDA's conclusion that these technologies are now mature, well-characterized, and manageable within the 510(k) framework.

The proposed rule would create a new regulatory category at 21 CFR 866.6075, titled "Nucleic Acid-Based Test Systems for Use with a Corresponding Approved Oncology Therapeutic Product." It affects tests currently approved under product codes OWD, PJG, PQP, and SFL -- approximately 35 original PMAs and 403 PMA supplements approved since 2011.

What this means for CDx developers:

• Dramatic cost reduction: PMA user fees in FY 2025 were approximately $541,000, compared to approximately $24,000 for a 510(k) -- roughly a 20-to-1 difference
• Faster review timelines: 510(k) reviews typically take 90 days versus 180+ days for PMA
• Predetermined Change Control Plans (PCCPs): Enable future modifications -- such as panel expansions and algorithm refinements -- without repeated full submissions
• Broader access: Lower barriers to entry may increase the number of available CDx, improving patient access to biomarker testing

Affected test types include:

• NGS panels for oncology
• PCR hotspot assays
• MSI/LOH signature detection
• Gene-specific mutation assays

The FDA justified the reclassification based on 14+ years of real-world safety data: 147 Medical Device Reports (more than 95% reported no clinical harm), 27 total recalls, and risk patterns deemed manageable through special controls rather than case-by-case PMA scrutiny.

Special controls under the proposed rule require:

• Analytical validity specifications (accuracy, precision, limit of detection, analytical sensitivity and specificity)
• Clinical validity demonstration using representative patient specimens
• Bioinformatics pipeline validation
• Clear design and labeling requirements consistent with the corresponding drug's FDA-approved labeling
• Quality system standards

Critical Limitation: Under the proposed rule, CDx labeling must be consistent with the corresponding therapeutic product's FDA-approved labeling. Diagnostic manufacturers cannot independently support drug-linked claims without corresponding therapeutic label updates. This creates a structural dependency: even when a biomarker-drug relationship is scientifically well-established, the CDx cannot make claims that exceed the approved drug label.

Important note: This proposed reclassification applies only to nucleic acid-based oncology CDx. Protein-based CDx (such as IHC assays for PD-L1 or HER2), non-oncology CDx, and novel biomarker tests without established predicates would continue to require PMA.

CDx Group Labeling (FDA 2020 Guidance)

In April 2020, the FDA finalized guidance on "Developing and Labeling In Vitro Companion Diagnostic Devices for a Specific Group or Class of Oncology Therapeutic Products." This guidance represents a significant shift from the traditional one-CDx-one-drug labeling model.

Under this framework, a CDx can be labeled for a group or class of oncology therapeutics rather than for each individual therapeutic product. For example, rather than labeling a BRAF V600E CDx separately for each BRAF inhibitor, the CDx can be labeled for the class of BRAF-targeted therapies. This approach:

• Reduces the number of PMA supplements required when new drugs in the same class are approved
• Simplifies CDx labeling for clinicians and laboratories
• Enables faster CDx availability for new drugs targeting established biomarkers
• Aligns the CDx labeling paradigm with the reality that many oncology biomarkers predict response to a class of agents, not just a single drug

Co-Development with Therapeutics

The co-development model is the gold standard for companion diagnostics. It aligns the diagnostic and therapeutic development programs from early clinical stages, ensuring that the CDx is validated using specimens from the same clinical trials that support the drug's efficacy claims.

The Drug-Diagnostic Co-Development Model

Co-development typically follows this trajectory:

Discovery and Early Development (Phase I): The therapeutic target and associated biomarker are identified. The diagnostic company begins developing a prototype assay. The drug and diagnostic sponsors enter a co-development agreement that defines roles, responsibilities, intellectual property, data sharing, and regulatory strategy.

Phase II Clinical Trials: The CDx prototype (or a clinical trial assay version) is used to select or stratify patients in Phase II. Analytical validation of the clinical trial assay begins. Data from Phase II inform the final CDx design, including cutoff values and scoring algorithms.

Phase III Pivotal Trials: The CDx (or a bridging-validated clinical trial assay) is used to identify the biomarker-positive population for enrollment in the pivotal trial. Specimens from pivotal trial subjects are tested with the final commercial CDx to generate clinical validation data. Bridging studies may be needed to demonstrate concordance between the clinical trial assay and the commercial CDx.

Regulatory Submission: The drug sponsor submits a New Drug Application (NDA) or Biologics License Application (BLA) to CDER or CBER. The CDx sponsor submits a PMA to CDRH. Cross-referencing letters are exchanged between the submissions. The reviews proceed in parallel, with coordination between review divisions.

Simultaneous Approval: Ideally, the drug and CDx are approved on the same day. The drug label references the specific CDx, and the CDx label references the corresponding therapeutic indication.

FDA Inter-Center Collaboration

The coordination between CDER (or CBER) and CDRH is governed by internal agreements and the 2016 guidance on co-development. Key features include:

• Pre-submission meetings: Joint pre-submission meetings can include representatives from both CDER/CBER and CDRH, allowing the drug and diagnostic sponsors to align regulatory expectations simultaneously
• Cross-referencing: The drug NDA/BLA references the CDx PMA, and vice versa, enabling each review team to access relevant data from the other submission
• Coordinated review timelines: FDA aims to synchronize review milestones so that both products can be approved concurrently
• Parallel review program: A formal program allowing coordinated review of therapeutics and their companion diagnostics under a single, harmonized timeline

Timing Considerations and Bridging Studies

One of the most challenging aspects of co-development is timing. The clinical trial assay used in Phase II and Phase III is often not the final commercial CDx. It may be a research-use-only (RUO) version, a laboratory-developed test, or a prototype that differs from the final manufactured product in reagents, platform, or scoring algorithm.

When differences exist between the clinical trial assay and the commercial CDx, bridging studies are required to demonstrate concordance. A bridging study tests a set of clinical specimens -- typically banked specimens from the pivotal trial -- on both the clinical trial assay and the commercial CDx, then compares results.

Practical Tip: Begin bridging study planning early. The most common cause of CDx approval delays is insufficient specimen availability for bridging studies. Ensure that your clinical trial protocol mandates collection and banking of sufficient tissue or blood specimens, with appropriate consent for diagnostic device development use. Specimen banking logistics should be defined in the co-development agreement.

EU IVDR Requirements for CDx

The European Union's In Vitro Diagnostic Regulation (EU 2017/746, IVDR) represents a fundamental shift in how companion diagnostics are regulated in Europe. Under the previous directive (IVDD), most IVDs were self-certified by manufacturers. Under the IVDR, companion diagnostics fall under significantly higher regulatory scrutiny.

Classification Under IVDR

The IVDR classifies IVDs into four risk classes: A (lowest risk), B, C, and D (highest risk). Companion diagnostics are classified under Rule 3 of Annex VIII, which assigns devices intended to be used as companion diagnostics to Class C.

However, this classification can be higher depending on the specific application:

IVDR Class	CDx Application Examples	Notified Body Involvement
Class C	Most CDx (e.g., HER2 testing for trastuzumab, EGFR testing for gefitinib)	Required -- conformity assessment by Notified Body
Class D	CDx used for blood screening, certain high-risk infectious disease testing	Required -- highest scrutiny, EU Reference Laboratory involvement

Class C and D devices require conformity assessment by an EU Notified Body -- a fundamental change from the self-certification model under the IVDD.

EMA Consultation Requirement

One of the most distinctive features of the IVDR framework for companion diagnostics is the mandatory consultation with the European Medicines Agency (EMA) or a National Competent Authority (NCA). Under Article 48(3)-(4) of the IVDR, the Notified Body assessing a CDx must seek a scientific opinion on the suitability of the companion diagnostic for the corresponding medicinal product.

This consultation process evaluates:

• The scientific validity of the biomarker in relation to the therapeutic product
• The analytical performance of the CDx as it pertains to the target medicine
• The clinical performance evidence supporting the CDx-therapeutic relationship

The consultation nominally takes 60 days but can be extended. The EMA or NCA opinion is not binding on the Notified Body, but it carries significant weight in the conformity assessment decision. This creates a multi-stakeholder review process involving the CDx manufacturer, the Notified Body, and the medicines regulatory authority -- a structure that has no direct equivalent in the FDA system, where CDRH handles the CDx review internally while coordinating with CDER or CBER.

Practical Note: The EMA consultation requirement adds both time and complexity to the EU CDx approval process. Manufacturers should budget additional months for this step and prepare a consultation dossier that clearly articulates the clinical evidence linking the CDx to the therapeutic product. Early engagement with the Notified Body on the consultation strategy is advisable.

Notified Body Capacity and Designation Challenges

The transition from IVDD to IVDR has created a significant bottleneck: the limited number of Notified Bodies designated under the IVDR. As of early 2026, only a small number of Notified Bodies have been designated for IVDR conformity assessment, and their capacity is heavily subscribed.

This capacity constraint has several implications for CDx manufacturers:

• Extended timelines: Wait times for Notified Body engagement can be 12 to 24 months or longer, depending on the complexity of the device and the Notified Body's workload
• Strategic Notified Body selection: Manufacturers must evaluate which designated Notified Bodies have expertise in molecular diagnostics, NGS, and companion diagnostic-specific requirements
• Prioritization decisions: Some manufacturers have been forced to prioritize EU market access for their highest-revenue CDx products and delay or abandon EU submissions for lower-volume tests
• Market access gaps: The combination of extended transition timelines and limited Notified Body capacity has raised concerns about potential gaps in CDx availability for European patients

The European Commission has acknowledged these challenges and implemented extended transition periods (discussed below), but the fundamental capacity issue remains a strategic consideration for any CDx manufacturer planning EU market entry.

Performance Evaluation Requirements

Under the IVDR, CDx manufacturers must compile a performance evaluation report that includes:

• Scientific validity: Evidence that the biomarker is associated with a clinical condition or physiological state relevant to the therapeutic product
• Analytical performance: Sensitivity, specificity, accuracy, precision, trueness, measuring range, limit of detection, limit of quantitation, linearity, cut-off values, interference, cross-reactivity
• Clinical performance: Evidence that the CDx results are associated with clinical outcomes in the intended patient population -- typically derived from clinical performance studies or, where justified, from existing clinical data and literature

The clinical evidence requirements under IVDR are substantially more rigorous than under IVDD. Manufacturers must demonstrate not just that the test accurately detects the biomarker, but that the biomarker test result is clinically meaningful in the context of the corresponding therapeutic.

Transition Timelines

The IVDR entered into application on May 26, 2022. However, recognizing the challenges facing the IVD industry and the limited capacity of Notified Bodies designated under IVDR, the European Commission implemented extended transition periods:

• Class D devices: Original deadline was May 2025; extended to May 2027 under certain conditions
• Class C devices (including most CDx): Extended deadline of May 2027 or May 2028, depending on whether a certificate was held under IVDD
• Class B and A sterile: Extended deadline of May 2028 or May 2029

Important: These transition timelines assume that manufacturers have submitted applications to a Notified Body before the applicable deadline. Manufacturers who have not engaged with a Notified Body by the relevant cutoff dates will not benefit from the extended timelines and their products may lose market access.

Post-Market Requirements

The IVDR introduces robust post-market surveillance obligations for CDx:

• Post-Market Surveillance Plan: Systematic collection and analysis of experience data
• Post-Market Performance Follow-Up (PMPF): Continuous active monitoring of clinical performance -- analogous to PMCF for medical devices under MDR
• Periodic Safety Update Reports (PSURs): Required for Class C and D devices, at least annually for the first two years after CE marking and at least every two years thereafter
• Serious Incident Reporting: Via the EUDAMED database, with defined timelines for reporting

FDA vs. IVDR: A Structural Comparison

The FDA and EU IVDR frameworks for companion diagnostics share the same scientific foundations -- both demand robust analytical and clinical evidence -- but diverge significantly in operational structure and regulatory burden:

Dimension	FDA (United States)	IVDR (European Union)
Reviewing body	CDRH (single agency)	Notified Body + EMA/NCA consultation
Classification	Class III (PMA) or proposed Class II for nucleic acid-based oncology CDx	Class C (most CDx) or Class D
Labeling linkage	CDx label cross-references specific therapeutic; drug label mandates CDx use	CDx label references therapeutic; drug-CDx linkage established through EMA consultation
Review coordination	CDRH coordinates directly with CDER/CBER	Notified Body must seek EMA/NCA opinion under Article 48
Post-market requirements	PMA annual reports; MDR reporting	PSURs, PMPF, EUDAMED incident reporting
Typical timeline	6-12 months (PMA); potentially faster under proposed 510(k)	12-24+ months including NB engagement and EMA consultation

With the FDA's proposed down-classification of nucleic acid-based oncology CDx to Class II, the operational divergence between the two jurisdictions is widening. The US pathway is becoming faster and less expensive, while the EU pathway remains resource-intensive. This asymmetry has strategic implications: many CDx manufacturers are now sequencing their regulatory filings to prioritize the US market and use FDA approval data to support subsequent EU submissions.

Global CDx Regulatory Frameworks Beyond the US and EU

The companion diagnostics regulatory landscape is increasingly global. While the FDA and EU IVDR frameworks are the most established, several other major markets have developed -- or are actively developing -- CDx-specific regulatory approaches.

Japan (PMDA/MHLW)

Japan has the most developed CDx regulatory framework in Asia. In December 2013, the Ministry of Health, Labour and Welfare (MHLW) issued "Technical Guidance on Development of In Vitro Companion Diagnostics and Corresponding Therapeutic Products," establishing the principle that CDx and therapeutic product applications should be submitted simultaneously.

Key features of the Japanese CDx framework:

• Classification categories: CDx can be classified as a companion diagnostic device, a standard IVD, or a Software as a Medical Device (SaMD/program medical device). The final classification is determined by PMDA/MHLW consultation.
• Simultaneous submission: PMDA encourages -- and in many cases expects -- concurrent submission of the CDx and the corresponding therapeutic product, mirroring the FDA's approach
• Registration timeline: Approximately 9 months for IVD/CDx registration through PMDA
• Reimbursement: Japan has an established CDx reimbursement scheme with near-universal coverage for approved CDx. For example, NGS panel tests like FoundationOne CDx receive specific reimbursement point values
• PMDA consultation: Consultation with PMDA is typically required to determine CDx classification and the appropriate registration route

China (NMPA)

China's CDx regulatory framework is still evolving but is rapidly becoming more structured:

• Classification: CDx are classified as Class III IVD products under NMPA's existing IVD and medical device regulations
• Local clinical data requirements: Applications must generally be supported by clinical trials conducted in China, which adds significant time and cost for international CDx manufacturers
• Registration timeline: Approximately 2 to 3 years, depending on whether a local clinical trial is needed
• Recent guidance: In June 2022, China published guidelines for co-developed CDx and anti-tumor drugs, and in April 2021, updated guidance on tumor CDx intended use statements
• Review structure: NMPA conducts independent reviews through separate centers for drugs and devices, creating coordination challenges similar to (but distinct from) the FDA's inter-center review process
• Market significance: China's growing oncology market and increasing adoption of precision medicine make it an increasingly important jurisdiction for CDx manufacturers, despite the regulatory complexity

South Korea (MFDS)

South Korea is developing dedicated CDx regulations:

• Current approach: CDx are classified as IVDs or medical devices under the Korean IVD Act
• Concurrent review: Article 6 of the Korean IVD Act provides for concurrent review of CDx and corresponding therapeutics
• Clinical performance: Article 7 addresses clinical performance testing requirements
• Reimbursement: Granted by the Korean Ministry of Health and Welfare (MOHW) according to the drug label
• Future development: New dedicated CDx regulations are planned, which may introduce more specific requirements

Australia (TGA)

The Therapeutic Goods Administration (TGA) has an established CDx framework:

• Classification: CDx are classified as Class 3 IVDs
• Development approach: TGA recommends simultaneous CDx-drug development with concurrent regulatory assessment, aligning with both the FDA and EU approaches
• Alignment: Australia's requirements align substantially with EU and FDA standards
• Planned developments: In 2026, the TGA plans to publish a list of approved IVD CDx products, increasing transparency in the Australian market

Regulatory Harmonization Efforts

Several international initiatives aim to harmonize CDx regulation across jurisdictions:

• International Medical Device Regulators Forum (IMDRF): Working groups have addressed IVD companion diagnostic terminology and regulatory approaches
• WHO IVD Prequalification Program: Provides assessment guidance for regulatory authorities that lack established CDx-specific procedures, particularly important for expanding CDx access in low- and middle-income countries
• ICH harmonization: While ICH primarily addresses pharmaceutical regulation, its guidelines on biomarker qualification and clinical trial design indirectly influence CDx development standards

Strategic Tip: For CDx with global market ambitions, design your analytical and clinical validation program to satisfy the most stringent requirements across all target jurisdictions simultaneously. Generating a single, comprehensive data package is more efficient than conducting jurisdiction-specific studies sequentially. Pay particular attention to China's local clinical data requirements, which may necessitate dedicated clinical sites and specimen collection in-country.

Analytical and Clinical Validation

Validation of a companion diagnostic is a two-part process: analytical validation (does the test accurately measure what it claims to measure?) and clinical validation (does the test result predict the clinical outcome of interest?).

Analytical Validation

Analytical validation demonstrates that the CDx reliably and accurately detects and measures the biomarker of interest. The key parameters are:

Sensitivity (Analytical): The ability of the test to detect the biomarker when present. For a genomic CDx, this might be the ability to detect a specific mutation at a defined allele frequency. For an IHC-based CDx, it is the ability to detect protein expression above a defined threshold.

Specificity (Analytical): The ability of the test to correctly identify specimens that do not harbor the biomarker. High analytical specificity means low false-positive rates.

Accuracy: The degree of closeness between the CDx result and the true value, often assessed by comparison to a reference method or orthogonal assay.

Precision: The degree of agreement among repeated measurements under specified conditions. Precision is assessed at multiple levels:

• Repeatability (within-run precision)
• Intermediate precision (within-laboratory, across runs, operators, and days)
• Reproducibility (across laboratories and sites)

Limit of Detection (LoD): The lowest amount of the biomarker that the test can reliably detect. For molecular CDx, this is often expressed as a minimum variant allele frequency (e.g., 5% VAF for a tumor mutation) or minimum copy number.

Limit of Quantitation (LoQ): The lowest amount of the biomarker that can be quantitatively determined with acceptable precision and accuracy.

Interfering Substances: Assessment of whether common substances in clinical specimens (hemoglobin, bilirubin, lipids, medications, DNA cross-contamination) affect test performance.

Clinical Validation

Clinical validation demonstrates that the CDx result is associated with the clinical outcome of interest -- typically therapeutic response, progression-free survival, or overall survival in patients treated with the corresponding therapeutic.

Clinical validation study design for CDx typically involves one or more of the following approaches:

• Prospective-retrospective design: Banked specimens from a completed prospective clinical trial are tested with the CDx, and results are correlated with known clinical outcomes. This is the most common design for co-developed CDx and is widely accepted by FDA.
• Prospective clinical study: A new prospective study in which patients are tested with the CDx and treated based on results. This design provides the strongest evidence but is time-consuming and expensive.
• Retrospective analysis: Historical data from clinical practice or registry databases are used to correlate CDx results with outcomes. This design is acceptable in some cases but must address potential biases inherent in retrospective data.

Regulatory Expectation: The FDA expects clinical validation data to demonstrate that the CDx identifies a population that derives meaningful clinical benefit from the therapeutic product. The clinical validation study should use the same (or a demonstrated equivalent) version of the CDx that will be commercially marketed. Bridging data are required if the clinical trial assay differs from the commercial CDx.

Clinical Utility: The Third Pillar of CDx Evidence

While analytical validation and clinical validation are the two core pillars of CDx evidence, a third concept -- clinical utility -- is increasingly emphasized by regulators, payers, and health technology assessment (HTA) bodies. Clinical utility answers the question: does the use of this CDx actually improve patient outcomes or clinical decision-making in practice?

Clinical utility is distinct from clinical validity. A CDx can be clinically valid (its results correlate with therapeutic outcomes in a controlled study) without having demonstrated clinical utility (real-world evidence that using the CDx leads to better treatment decisions, improved survival, reduced adverse events, or lower healthcare costs compared to not using the CDx).

Why clinical utility matters:

• Payer coverage decisions: CMS and private payers increasingly require evidence of clinical utility -- not just analytical and clinical validity -- as a prerequisite for coverage. A CDx that is analytically perfect but does not demonstrably change clinical outcomes may face reimbursement challenges.
• HTA assessments: In the EU and other markets with HTA-driven reimbursement, health technology assessment bodies evaluate whether the CDx-therapeutic combination provides sufficient health economic value. Clinical utility evidence is central to these assessments.
• FDA's evolving expectations: While the FDA's CDx guidance focuses on analytical and clinical validation, the agency increasingly considers clinical utility in the context of breakthrough device designation and benefit-risk assessments.
• Professional guideline inclusion: Organizations like NCCN and ESMO consider clinical utility when recommending specific CDx in their treatment guidelines.

Approaches to demonstrating clinical utility include:

• Randomized controlled trials comparing CDx-guided therapy to standard-of-care (the gold standard, but rarely feasible solely for the diagnostic)
• Prospective-retrospective analyses showing that CDx-positive patients derive greater therapeutic benefit than CDx-negative patients
• Real-world evidence studies demonstrating improved outcomes in clinical settings where CDx is systematically used
• Health economic modeling showing cost-effectiveness of CDx-guided treatment strategies

Specimen Types and Pre-Analytical Variables

CDx performance can be significantly affected by pre-analytical variables -- factors that occur before the analytical measurement itself. These include:

• Specimen type: Formalin-fixed paraffin-embedded (FFPE) tissue, fresh-frozen tissue, whole blood, plasma, bone marrow, cerebrospinal fluid
• Fixation conditions: Fixation time, fixative type, and fixation delay can dramatically affect IHC and molecular assay performance
• Specimen age: Degradation of nucleic acids and proteins over time in archived FFPE blocks
• Tumor content: Minimum tumor cellularity required for accurate mutation detection
• DNA/RNA quality and quantity: Input requirements vary by platform
• Collection tube type: For liquid biopsy, the choice of blood collection tube affects cfDNA yield and integrity

CDx labeling must specify acceptable specimen types and pre-analytical conditions. Validation studies must demonstrate acceptable performance across the range of pre-analytical variables that the test is intended to accommodate.

Key Biomarker Types in CDx

Companion diagnostics target a range of biomarker types, each with distinct biological, analytical, and regulatory characteristics.

Genomic Biomarkers

Genomic biomarkers -- DNA-level alterations -- are the most common targets for companion diagnostics in oncology:

Biomarker	Alteration Type	Therapeutic Association	CDx Technology
HER2 (ERBB2)	Gene amplification / protein overexpression	Trastuzumab, pertuzumab, T-DXd	IHC, FISH, ISH, NGS
EGFR	Activating mutations (exon 19 del, L858R, T790M, exon 20 insertions)	Osimertinib, erlotinib, gefitinib	PCR, NGS
BRAF V600E/K	Point mutation	Vemurafenib, dabrafenib, encorafenib	PCR, NGS
ALK	Gene rearrangement / fusion	Crizotinib, alectinib, lorlatinib	FISH, IHC, NGS
KRAS G12C	Point mutation	Sotorasib, adagrasib	PCR, NGS
BRCA1/2	Germline and somatic mutations	Olaparib, rucaparib, niraparib	NGS, PCR
ROS1	Gene rearrangement / fusion	Crizotinib, entrectinib	FISH, NGS
NTRK	Gene fusions	Larotrectinib, entrectinib	NGS, FISH, IHC
RET	Gene fusions and mutations	Selpercatinib, pralsetinib	NGS
MSI-H/dMMR	Microsatellite instability / mismatch repair deficiency	Pembrolizumab (tumor-agnostic)	PCR, IHC, NGS

Protein Biomarkers

Protein-level biomarkers are typically detected by immunohistochemistry (IHC) or immunoassays:

PD-L1: Expressed on tumor cells and immune cells, PD-L1 is detected by IHC using clone-specific antibodies. Different therapeutic products use different PD-L1 CDx with different scoring algorithms and cutoffs, creating significant complexity for pathology laboratories. The PD-L1 IHC 22C3 pharmDx (Dako/Agilent) is the CDx for pembrolizumab; the PD-L1 IHC 28-8 pharmDx is the complementary diagnostic for nivolumab.

HER2 protein: Detected by IHC (0, 1+, 2+, 3+ scoring) and confirmed by in situ hybridization (ISH) for equivocal (2+) cases. The advent of antibody-drug conjugates like trastuzumab deruxtecan (T-DXd) has expanded the relevant HER2 population to include HER2-low (IHC 1+ or 2+/ISH-negative) tumors, requiring updates to CDx labeling and potentially new CDx assays.

ctDNA and Liquid Biopsy CDx

Liquid biopsy companion diagnostics analyze circulating tumor DNA (ctDNA) or other circulating biomarkers in blood. This represents one of the fastest-growing areas in CDx development:

• Non-invasive sampling: Blood draw instead of tissue biopsy
• Real-time monitoring: Can detect resistance mutations and disease progression before radiographic changes
• Tumor heterogeneity: ctDNA can represent multiple tumor sites, potentially capturing heterogeneity missed by single-site tissue biopsy
• Challenges: Lower sensitivity than tissue-based testing (particularly at low tumor burden), pre-analytical variability in cfDNA handling, clonal hematopoiesis of indeterminate potential (CHIP) as a source of false positives

NGS-Based CDx Panels

Next-generation sequencing panels represent the most significant technological evolution in companion diagnostics. Rather than testing for a single biomarker, NGS CDx panels simultaneously assess dozens or hundreds of genes, identifying patients for multiple therapeutic options in a single test.

FDA-approved NGS CDx panels include:

• FoundationOne CDx (Foundation Medicine / Roche): Tissue-based 324-gene panel; approved as CDx for multiple targeted therapies across tumor types
• Guardant360 CDx (Guardant Health): Liquid biopsy 74-gene panel; approved as CDx for osimertinib (EGFR) and other indications
• MSK-IMPACT (Memorial Sloan Kettering): 468-gene panel; authorized via the FDA's tumor profiling pathway
• Oncomine Dx Target Test (Thermo Fisher): Tissue-based panel for NSCLC biomarkers
• Praxis Extended RAS Panel (Illumina): For RAS mutation detection in colorectal cancer

Notable CDx Case Studies

HER2/Herceptin: The First Companion Diagnostic

The approval of trastuzumab (Herceptin) in 1998 for HER2-positive metastatic breast cancer, paired with the HercepTest IHC assay, established the companion diagnostic paradigm. The pivotal trial (H0648g) demonstrated that trastuzumab added to chemotherapy significantly improved overall survival -- but only in patients whose tumors overexpressed HER2. The clinical trial used both IHC and FISH testing to identify HER2-positive patients.

The HER2/Herceptin story demonstrated several principles that continue to define CDx development: the need for robust biomarker selection, the importance of validating the CDx in the same trial population that supports the drug's efficacy, and the reality that CDx development can take as long as drug development itself.

PD-L1/Pembrolizumab: The Complexity of Scoring Systems

The development of PD-L1 CDx for immune checkpoint inhibitors exposed the challenge of biomarker complexity. Different checkpoint inhibitors use different PD-L1 antibody clones, different scoring systems (Tumor Proportion Score, Combined Positive Score, Immune Cell Score), and different positivity cutoffs. This created a situation where a patient's PD-L1 status could depend on which CDx was used, which antibody clone was applied, and which scoring algorithm was interpreted.

FDA addressed this partly by approving the 22C3 pharmDx assay as the CDx for pembrolizumab with indication-specific scoring algorithms and cutoffs. But the broader issue -- the need for biomarker harmonization and potential interchangeability of CDx assays -- remains an active area of discussion.

BRCA/Olaparib: Germline and Somatic Testing

The approval of olaparib (Lynparza) for BRCA-mutated ovarian cancer introduced the concept of CDx testing for both germline and somatic mutations. The BRACAnalysis CDx (Myriad Genetics) was initially approved to test germline BRCA mutations. Later, FoundationOne CDx was approved as a CDx to detect both germline and somatic BRCA mutations, broadening the eligible patient population.

This case study illustrates the evolution from single-analyte CDx to panel-based CDx, and the regulatory pathway for expanding CDx indications to encompass new mutation types (germline to somatic) and new therapeutic indications.

KRAS G12C: A Previously Undruggable Target

For decades, KRAS was considered "undruggable." The development of sotorasib (Lumakras) and adagrasib (Krazati) for KRAS G12C-mutated non-small cell lung cancer represented a breakthrough in both therapeutics and diagnostics. The KRAS G12C mutation is present in approximately 13% of NSCLC adenocarcinomas, and identifying these patients requires molecular testing -- either single-gene PCR or NGS panel testing.

The CDx for sotorasib was the therascreen KRAS RGQ PCR Kit (QIAGEN), while FoundationOne CDx also received approval as a CDx for this indication. The KRAS G12C case demonstrates how a single biomarker can have multiple approved CDx across different technology platforms.

FoundationOne CDx: The Pan-Tumor NGS CDx

FoundationOne CDx (Foundation Medicine / Roche) was approved in 2017 as the first comprehensive genomic profiling test to receive FDA approval as a CDx. It tests 324 genes for substitutions, insertions/deletions, copy number alterations, and rearrangements, plus genomic signatures including microsatellite instability (MSI) and tumor mutational burden (TMB).

What makes FoundationOne CDx distinctive is its pan-tumor, multi-biomarker approach. A single test can serve as the CDx for multiple therapeutic products across multiple tumor types. As of 2026, FoundationOne CDx is approved as a CDx for more than 20 therapeutic indications across multiple cancer types. This model -- one test, many therapies -- represents the future direction of companion diagnostics.

Guardant360 CDx: Liquid Biopsy Enters the CDx Space

Guardant360 CDx (Guardant Health) was the first liquid biopsy test to receive FDA approval as a companion diagnostic. Approved initially in 2020 for EGFR mutation detection in NSCLC (as a CDx for osimertinib), it demonstrated that blood-based testing could meet the analytical and clinical validation standards required for CDx approval.

The approval of Guardant360 CDx validated the liquid biopsy approach for CDx applications and opened the door for blood-based companion diagnostics across multiple indications. It also raised important questions about the concordance between tissue-based and liquid biopsy CDx results, and about the appropriate use of liquid biopsy when tissue is available.

NGS-Based Companion Diagnostics

Next-generation sequencing has fundamentally changed the companion diagnostic landscape. Instead of developing a separate test for each biomarker, NGS panels can test for dozens or hundreds of biomarkers simultaneously, enabling a single test to serve as the CDx for multiple therapeutic products.

FDA Framework for NGS CDx

The FDA has developed a specific regulatory framework for NGS-based companion diagnostics, recognizing that the traditional one-test-one-drug model does not accommodate panel-based testing. Key elements include:

• Tumor profiling authorization: FDA has created mechanisms for approving NGS panels as CDx for multiple concurrent therapeutic indications
• Modular PMA supplements: When a new therapeutic indication is added to an existing NGS CDx panel, the manufacturer can submit a PMA supplement that leverages previously reviewed analytical validation data, adding only the new clinical validation evidence
• Variant-level validation: FDA expects validation at the variant level, not just the gene level. For a mutation detected by the panel, the manufacturer must demonstrate that the specific variant is detectable with defined analytical performance

Pan-Cancer CDx Concept

The concept of a pan-cancer companion diagnostic -- a single test approved across all solid tumor types -- emerged with the approval of tumor-agnostic therapies. Pembrolizumab's approval for MSI-H/dMMR solid tumors (regardless of cancer type) and larotrectinib's approval for NTRK fusion-positive solid tumors created a need for CDx that could be applied across all tumor types.

This model requires:

• Analytical validation across multiple specimen types and tumor histologies
• Clinical validation spanning multiple cancer types (often through basket trial designs)
• Labeling that accommodates tumor-agnostic indications

Challenges in NGS CDx Validation

NGS-based CDx face unique validation challenges:

• Variant diversity: A single gene may harbor hundreds of clinically relevant variants, and the panel must detect them all reliably
• Bioinformatics pipeline validation: The variant calling algorithms, alignment algorithms, and quality filters must be validated as rigorously as the wet-lab chemistry
• Reference materials: Availability of validated reference materials for rare variants is limited; the Microarray and Sequencing Quality Control (MAQC/SEQC) consortium and organizations like the National Institute of Standards and Technology (NIST) have developed reference materials, but gaps remain
• Variant classification: Determining whether a detected variant is pathogenic, likely pathogenic, of uncertain significance, or benign requires curated databases and expert interpretation
• Turnaround time: NGS workflows are inherently longer than single-analyte PCR tests, which can affect treatment timelines

Practical Tip: When developing an NGS-based CDx, begin bioinformatics pipeline validation early and treat it with the same rigor as wet-lab validation. FDA expects documentation of the entire computational pipeline, including alignment algorithms, variant callers, quality metrics, and filters. Changes to the bioinformatics pipeline may require PMA supplements.

Complementary vs. Companion Diagnostics: A Deeper Look

The distinction between companion and complementary diagnostics has significant implications for drug labeling, reimbursement, and clinical practice.

Regulatory Classification

Feature	Companion Diagnostic (CDx)	Complementary Diagnostic
Regulatory requirement	Essential -- drug label mandates test use	Informational -- drug can be used without test
Drug labeling language	"Test required" or "indicated for patients with [biomarker] as determined by [specific CDx]"	"Information about [biomarker] may inform treatment decisions"
Regulatory pathway	PMA (most common)	PMA or 510(k)
Drug approval linkage	Drug and CDx approved contemporaneously	Test approval may be independent of drug approval
Clinical impact of not testing	Patient may not receive appropriate treatment	Physician loses one data point but can still prescribe

The PD-L1 Example in Detail

The PD-L1 testing landscape illustrates the companion/complementary distinction perfectly:

Pembrolizumab in first-line NSCLC (single agent): PD-L1 IHC 22C3 pharmDx is a companion diagnostic. Patients must have PD-L1 TPS greater than or equal to 1% (or greater than or equal to 50% for single-agent first-line use in certain indications). The test is required.

Nivolumab in second-line NSCLC: PD-L1 IHC 28-8 pharmDx was approved as a complementary diagnostic. PD-L1 expression provides prognostic information, but nivolumab is indicated regardless of PD-L1 status. Physicians may use the PD-L1 result to inform discussions with patients, but the test is not a prerequisite for treatment.

Atezolizumab in urothelial carcinoma: The VENTANA PD-L1 (SP142) assay was approved as a complementary diagnostic using an Immune Cell (IC) scoring algorithm -- a different antibody clone and scoring method than either the 22C3 or 28-8 assays.

This fragmentation -- different antibodies, different scoring systems, different regulatory classifications for the same biomarker class -- remains one of the most challenging aspects of CDx for pathology laboratories and clinicians.

Commercial and Reimbursement Considerations

Developing a companion diagnostic that achieves regulatory approval is only half the battle. Securing reimbursement and achieving broad market adoption require a separate but equally important strategy.

CMS Coverage

In the United States, the Centers for Medicare and Medicaid Services (CMS) is the largest single payer and sets coverage policies that influence the broader market:

National Coverage Determinations (NCDs): CMS can issue national coverage determinations for specific categories of tests. The NCD for NGS in advanced cancer (NCD 90.2) covers FDA-approved NGS CDx tests for patients with advanced cancer who are considering treatment, regardless of cancer type. This NCD was a landmark decision that established broad national coverage for NGS-based CDx.

Local Coverage Determinations (LCDs): In the absence of an NCD, Medicare Administrative Contractors (MACs) make local coverage decisions. LCD coverage for CDx varies by region and by specific test.

Medicare Coverage of Innovative Technology (MCIT): The MCIT pathway, which was proposed to provide automatic Medicare coverage for FDA breakthrough-designated devices, has had a complex regulatory history. As of 2026, CMS has refined the MCIT framework, and breakthrough-designated CDx may benefit from expedited coverage pathways.

Coding and Payment

CDx reimbursement depends on proper coding:

• CPT codes: The American Medical Association's CPT code system includes specific codes for molecular pathology tests, IHC, ISH, and NGS panels. Multi-analyte CDx panels may use Proprietary Laboratory Analyses (PLA) codes.
• HCPCS codes: Used for Medicare billing; CDx administered in hospital outpatient settings may use specific HCPCS codes.
• AMA PLA codes: For proprietary tests that do not fit existing CPT codes, PLA codes provide a pathway for unique test-specific coding.

Hospital vs. Reference Laboratory Adoption

CDx adoption patterns vary by testing setting:

Hospital/Academic Medical Center laboratories prefer CDx platforms that integrate into existing workflows, offer reasonable turnaround times, and are supported by pathologist interpretation. IHC and FISH-based CDx are well-suited to hospital laboratories.

Reference laboratories (e.g., Foundation Medicine, Guardant Health, Myriad Genetics) specialize in complex molecular testing, including NGS panels and liquid biopsy. Many CDx are practically accessible only through reference laboratories due to the capital equipment, bioinformatics infrastructure, and expertise required.

Market Access Tip: A CDx reimbursement strategy should be developed in parallel with clinical development, not after approval. Engage with payers early, generate health economic data, and secure coding and coverage before or concurrent with FDA approval. The most analytically elegant CDx will fail commercially if it lacks adequate reimbursement.

Market Access Strategy Framework

Phase	Key Activities	Timeline Relative to Approval
Early planning	Identify target payer landscape; begin health economic modeling	3-4 years before approval
Evidence generation	Design clinical trials to capture economic endpoints; begin payer advisory boards	2-3 years before approval
Pre-launch	Submit coding applications (CPT/PLA); engage with MACs; prepare coverage dossiers	12-18 months before approval
Launch	Execute payer contracts; establish reference lab partnerships; train hospital labs	At approval
Post-launch	Monitor coverage decisions; address denials; expand covered indications	Ongoing

Emerging Trends in Companion Diagnostics

AI/ML in CDx

Artificial intelligence and machine learning are entering the companion diagnostic space in several ways:

• Digital pathology with AI scoring: AI algorithms can standardize PD-L1 scoring and HER2 interpretation, reducing inter-observer variability among pathologists
• AI-assisted variant interpretation: Machine learning models can assist in classifying variants of uncertain significance detected by NGS CDx panels
• Computational biomarkers: AI models that predict treatment response from routine histology images (H&E stains) without a molecular assay, potentially serving as pre-screening tools before CDx testing
• Multimodal CDx: Combining genomic, proteomic, and imaging data using AI to create composite biomarker scores with greater predictive power than any single analyte

Multi-Cancer Early Detection (MCED)

While not traditional CDx, multi-cancer early detection tests (such as Galleri by GRAIL) represent a convergence of diagnostic technology and therapeutic decision-making. As MCED tests mature and are paired with specific therapeutic interventions, some may evolve into companion diagnostics for cancer interception strategies.

Minimal Residual Disease (MRD)

MRD testing -- detecting trace amounts of cancer after treatment -- is emerging as a companion diagnostic application for maintenance therapy and treatment intensification decisions. ctDNA-based MRD assays are being co-developed with therapeutic products in colorectal cancer, lung cancer, and hematologic malignancies.

Pharmacogenomics CDx

Beyond oncology, companion diagnostics are expanding into pharmacogenomics -- testing for germline genetic variants that affect drug metabolism, efficacy, or toxicity. Examples include:

• CYP2D6 testing for eligibility for certain pain medications
• HLA-B*5701 testing before abacavir use in HIV
• UGT1A1 testing for irinotecan dosing
• DPYD testing for fluoropyrimidine toxicity risk

CDx in Cell and Gene Therapy

The rapid growth of cell and gene therapies -- CAR-T cells, gene editing, viral vector therapies -- is creating new CDx requirements. These therapies often require patient selection based on specific antigen expression (e.g., CD19 for tisagenlecleucel), disease genotype, or other biomarkers. CDx for cell and gene therapy face unique challenges, including small patient populations, rapidly evolving therapeutic landscapes, and the need for near-real-time testing to support manufacturing timelines.

CDx for Antibody-Drug Conjugates (ADCs)

The ADC class has created new CDx challenges. Trastuzumab deruxtecan (T-DXd), for instance, has shown efficacy not only in traditional HER2-positive breast cancer but also in HER2-low tumors. This has required rethinking HER2 CDx scoring, as the traditional binary positive/negative classification no longer captures the clinically relevant population. IHC scoring nuances, particularly the distinction between IHC 0 and IHC 1+ (previously considered clinically irrelevant), have become treatment-defining.

Diversity, Equity, and Inclusion in CDx Development

A growing concern in the CDx field is the lack of diversity in the genomic research and clinical trial populations that underpin companion diagnostic development. Most large-scale genomic databases and pivotal clinical trials have been conducted predominantly in populations of European descent. This creates several risks:

• Biomarker prevalence differences: The prevalence of specific mutations can vary significantly across ethnic and racial groups. EGFR mutations, for example, are present in approximately 10-15% of NSCLC patients of European descent but in 40-55% of East Asian NSCLC patients. CDx cutoffs and scoring algorithms validated primarily in one population may not perform optimally in others.
• Variant classification gaps: Genomic databases used for variant interpretation may have fewer entries for variants found predominantly in underrepresented populations, leading to higher rates of variants of uncertain significance (VUS) in these groups.
• Clinical validation limitations: If the pivotal clinical trial population does not reflect the real-world patient population, the clinical validation evidence for the CDx may not fully support its use across all demographic groups.
• Health equity implications: If CDx are less well-validated in certain populations, patients in those groups may receive less accurate biomarker testing, potentially leading to inappropriate treatment decisions.

Regulatory agencies, professional organizations, and industry groups are increasingly calling for more diverse and representative clinical trial populations in CDx development programs. The FDA's guidance on diversity plans for clinical trials, finalized in 2024, applies to both therapeutics and their companion diagnostics.

CDx Intellectual Property and Business Models

The intellectual property landscape for companion diagnostics involves multiple overlapping rights and complex business relationships:

• Patent protection: CDx may be protected by patents covering the biomarker itself, the detection method, specific reagents or probes, scoring algorithms, or the combination of diagnostic result and therapeutic indication. The intersection of diagnostic and therapeutic patents creates a complex IP landscape.
• Licensing agreements: CDx development often requires licensing access to biomarker IP held by academic institutions, pharmaceutical companies, or other diagnostic companies. License terms -- including exclusivity, field-of-use restrictions, and royalty rates -- significantly affect CDx commercial viability.
• Co-development agreements: The legal framework governing the relationship between the pharmaceutical sponsor and the CDx developer is critical. Key terms include intellectual property ownership and licensing, data sharing and cross-referencing rights, exclusivity periods, commercialization responsibilities, and termination rights.
• Business models: CDx business models vary widely, from the traditional model (pharmaceutical company partners with a single CDx developer for exclusive co-development) to the platform model (a single NGS panel serves as CDx for multiple therapeutics from multiple pharmaceutical companies, as with FoundationOne CDx). The platform model is increasingly dominant, as it offers economies of scale and broader clinical utility.

CDx Market Landscape

The global companion diagnostics market has grown substantially and continues to expand:

• The CDx market is estimated at approximately $8-10 billion as of 2025-2026, with projections for continued double-digit growth driven by the expanding pipeline of targeted therapies and immuno-oncology agents
• As of 2026, there are more than 160 FDA-approved combination therapies that use companion diagnostics
• Oncology dominates the CDx market, accounting for approximately 70-80% of all approved CDx, but applications in infectious disease, cardiology, and rare diseases are growing
• Technology trends are shifting from single-analyte tests toward comprehensive NGS panels and liquid biopsy platforms, reflecting the clinical preference for obtaining maximum biomarker information from a single test
• Key market players include Roche/Foundation Medicine, Agilent/Dako, Guardant Health, Myriad Genetics, Thermo Fisher Scientific, Illumina, and QIAGEN

Before the era of CDx, drug efficacy rates in broad, unselected populations ranged from approximately 50-60%. With companion diagnostic-guided patient selection, response rates have more than doubled in some indications -- reaching above 90% for certain biomarker-defined populations. This improvement in therapeutic precision is the fundamental value proposition driving CDx adoption and investment.

Frequently Asked Questions

What is the difference between a companion diagnostic and a biomarker test?

A biomarker test is any assay that measures a biological marker. A companion diagnostic is a specific, FDA-approved biomarker test that is essential for the safe and effective use of a corresponding therapeutic product. All companion diagnostics are biomarker tests, but not all biomarker tests are companion diagnostics. The distinction is regulatory: a CDx has been reviewed and approved by FDA specifically for use with a named therapeutic product, and the therapeutic's labeling mandates its use.

Can an LDT be used instead of an FDA-approved CDx?

Technically, a physician can exercise clinical judgment in ordering any test, including an LDT. However, when a drug label specifically requires an FDA-approved CDx, using an LDT creates regulatory, liability, and reimbursement risks. The LDT may not have undergone the same level of analytical and clinical validation, and payers may deny claims for the therapeutic if the specified CDx was not used. With the FDA's 2024 final rule on LDT regulation, the regulatory landscape is shifting further toward requiring FDA oversight of tests used for treatment decisions.

How long does it take to develop a companion diagnostic?

CDx development typically takes 3 to 7 years, depending on the complexity of the biomarker, the technology platform, and the co-development timeline with the therapeutic product. Analytical validation alone typically takes 12 to 18 months. Clinical validation depends on the availability of clinical trial specimens and the timeline of the therapeutic's clinical development program. The regulatory submission and review process adds another 12 to 24 months.

What does a CDx PMA submission include?

A CDx PMA submission typically includes: device description and intended use, manufacturing information and quality system documentation, non-clinical (bench) testing data, analytical validation data (sensitivity, specificity, precision, accuracy, LoD, interfering substances), clinical validation data (correlation of CDx results with clinical outcomes from therapeutic clinical trials), labeling (including cross-reference to the corresponding therapeutic), and a summary of safety and effectiveness. The submission cross-references the therapeutic product's NDA or BLA.

Can one CDx be used for multiple drugs?

Yes. NGS-based CDx panels like FoundationOne CDx are approved as companion diagnostics for multiple therapeutic products across multiple tumor types. Each drug-indication pairing requires its own clinical validation evidence, but the analytical validation of the underlying platform and individual variant detection can be shared across indications. This multi-indication CDx model is increasingly preferred by oncologists and payers because it provides comprehensive biomarker information from a single test.

What is the cost of obtaining FDA approval for a CDx?

The total cost varies widely depending on the technology, the number of indications, and the scope of validation required. Rough estimates include: FDA user fees ($500K to $2M+ for PMA), analytical validation studies ($500K to $2M), clinical validation studies ($1M to $10M+, depending on specimen availability and trial design), manufacturing and quality system development ($500K to $2M), and regulatory consulting and submission preparation ($300K to $800K). Total costs for a novel CDx PMA can range from $5M to $20M or more.

How is a CDx labeled differently from a standard IVD?

CDx labeling must include specific cross-references to the corresponding therapeutic product, including the drug name, approved indications, and the specific patient population identified by the CDx result. The intended use statement must reference the therapeutic product. The labeling must include the clinical validation data linking CDx results to therapeutic outcomes. Standard IVDs do not include these therapeutic cross-references.

What happens if a CDx is not available when a drug is approved?

The FDA strongly prefers simultaneous approval of the drug and CDx. However, in rare cases, a drug may be approved without an approved CDx -- particularly under accelerated approval pathways or when the clinical benefit is so substantial that delaying drug approval to wait for CDx approval would be contrary to the public interest. In these cases, the drug label may reference a specific biomarker requirement without naming a specific CDx, and the FDA typically expects a CDx to follow within a defined timeframe. This situation creates significant clinical and regulatory complexity and is not the preferred model.

Are companion diagnostics required outside the United States?

Requirements vary by jurisdiction. In the EU, the IVDR classifies CDx as Class C (or higher) devices requiring Notified Body conformity assessment, but the regulatory linkage between CDx and therapeutic approval is less prescriptive than in the US. In Japan, the PMDA has a CDx regulatory framework that mirrors the FDA's approach in many respects. In China, NMPA has been developing CDx-specific guidance. In many other markets, CDx may be registered as standard IVDs without a formal regulatory linkage to the therapeutic product, though this is changing as precision medicine adoption grows globally.

What role do professional guidelines play in CDx adoption?

Professional guidelines from organizations like the National Comprehensive Cancer Network (NCCN), the College of American Pathologists (CAP), the Association for Molecular Pathology (AMP), and the European Society for Medical Oncology (ESMO) play a critical role in CDx adoption. These guidelines recommend specific biomarker tests for specific clinical scenarios and often specify whether an FDA-approved CDx is required or whether alternative testing (including LDTs) is acceptable. Inclusion in NCCN guidelines, in particular, is often a prerequisite for broad payer coverage and clinical adoption in the United States.

---

Companion diagnostics sit at the intersection of diagnostics, therapeutics, and regulatory science. They are the mechanism through which precision medicine delivers on its promise: matching the right patient to the right therapy based on objective, validated biomarker evidence. The regulatory frameworks in the US and EU are mature but continue to evolve, driven by advances in NGS, liquid biopsy, AI, and the expanding scope of targeted therapies. For manufacturers, the path to CDx approval requires early strategic planning, deep regulatory expertise, robust analytical and clinical validation, and a commercial strategy that addresses reimbursement and market access from the outset. For clinicians and patients, companion diagnostics represent one of the most consequential advances in modern medicine -- transforming treatment decisions from empirical to evidence-based, one biomarker at a time.