BIOPHARMAWIRE  |  RESOURCE LIBRARY

CDx Co-Development Guide

Companion diagnostic planning, regulatory requirements, co-development strategy, and the FDA/CDRH approval process

Updated Q4 2025  |  BioPharmaWire Editors  |  Based on FDA guidance: In Vitro Companion Diagnostic Devices (2014), Co-development of Companion Diagnostics (2023 draft)

A companion diagnostic (CDx) is an in vitro diagnostic (IVD) device that provides information essential for the safe and effective use of a corresponding therapeutic product. When a drug’s approval is contingent on a biomarker — identifying which patients will benefit, which will not respond, or which are at risk of serious adverse events — FDA requires a co-approved CDx as a condition of the drug’s marketing authorisation. As precision medicine has moved from concept to standard of care, CDx co-development has become a routine requirement for oncology programs and is increasingly common in rare disease, neurology, and immunology.

This guide covers the regulatory framework for CDx co-development, the practical co-development timeline aligned with drug clinical phases, partnership and contracting considerations, and the parallel submission and review process at FDA.

When is a CDx Required?

Mandatory CDx

FDA requires a co-approved CDx when: the drug label includes a biomarker-defined indication (e.g. ‘for patients with EGFR exon 19 deletions or exon 21 L858R substitutions’); clinical trial enrolment for the pivotal trial was restricted to biomarker-positive patients; or a specific patient subgroup must be identified before treatment to avoid a serious safety risk. In these cases, the drug cannot be approved without a simultaneously or previously approved CDx for the same biomarker in the same specimen type.

Complementary Diagnostic

A complementary diagnostic is a test that provides information that is clinically meaningful but not required for safe and effective use of the drug. The drug label may reference the test but treatment is not restricted to test-positive patients. The regulatory bar for complementary diagnostics is lower than for CDx — they do not require PMA approval and can be commercially available as laboratory-developed tests (LDTs) or 510(k)-cleared IVDs.

The Biomarker Classification Decision

The decision of whether a biomarker requires a mandatory CDx, qualifies as a complementary diagnostic, or needs no IVD at all is one of the most consequential early decisions in a precision medicine development program. It should be made in consultation with FDA before Phase 2 design is finalised. A biomarker that starts as exploratory in Phase 1 may become mandatory in Phase 3 if the Phase 2 data shows a strong predictive effect — and failing to plan the CDx co-development at that inflection point creates significant timeline risk.

The Regulatory Framework: FDA CDER/CDRH Co-Review

Two Agencies, One Approval

CDx co-development involves two FDA centres simultaneously: CDER (or CBER for biologics) reviews the drug; CDRH reviews the CDx device. The two centres coordinate on cross-labelling language, clinical validity standards, and approval timing. The result is a simultaneous or near-simultaneous approval of the drug and its CDx — the two products reference each other in their respective labels.

CDx Regulatory Pathway: PMA vs 510(k)

Most novel CDx devices require a Pre-Market Approval (PMA) application through CDRH — the highest regulatory standard for IVD devices, analogous to an NDA/BLA. PMA requires demonstration of safety and effectiveness through analytical validation (precision, accuracy, reproducibility) and clinical validation (that the test correctly identifies patients who benefit from the drug). A 510(k) pathway (substantial equivalence to a predicate device) is available when a predicate CDx for a similar biomarker exists — for example, a second KRAS G12C assay may use the first approved assay as a predicate.

CLIA and Laboratory Certification

All CDx tests used in the US must be performed in CLIA-certified laboratories. This is a practical consideration for drug sponsors: the CDx partner must ensure that laboratory sites participating in Phase 3 clinical trials are CLIA-certified and that the CDx assay is available for commercial use in CLIA-certified settings at launch. For complex assays (NGS panels, for example), the number of CLIA-certified laboratories that can perform the test at launch is a meaningful commercial constraint.

CDx Co-Development Timeline by Phase

Note:  The table below maps CDx co-development activities against the drug development timeline. Activities in grey italics are not yet required at that phase but should be planned. The most common failure mode in CDx programs is starting CDx co-development too late — assay lock before Phase 3 enrolment is a regulatory requirement, not an aspiration.

Development Phase	Drug Sponsor Activity	CDx Partner Activity	Joint / Regulatory Actions
PRE-CLINICAL & EARLY CLINICAL (IND FILING TO PHASE 1 COMPLETION)
IND Filing	Identify candidate biomarker(s); define hypothesis for patient selection	Not yet formally engaged; CDx partner identified informally	Early FDA/CDER consultation on biomarker strategy; no formal IVD regulatory action required at IND stage
Phase 1 (FIH)	Collect biospecimens; run exploratory biomarker analyses; identify sensitive/resistant subpopulations	Provide research-use-only (RUO) assay versions for Phase 1 biomarker work; begin assay feasibility	Agree on formal CDx partnership; execute co-development agreement; define assay development plan
PHASE 2 — ASSAY LOCK AND CLINICAL VALIDATION
Phase 2 design	Decide: biomarker-selected vs all-comers with retrospective analysis; define primary population	Develop investigational use only (IUO) assay version; begin analytical validation studies (sensitivity, specificity, reproducibility, precision)	Joint FDA meeting (Type B): align on CDx strategy, biomarker-positive definition, and Phase 3 design. Obtain FDA agreement on prospective vs retrospective validation approach
Phase 2 conduct	Lock biomarker-positive definition before Phase 3 enrolment. Collect concordance data between assay platforms.	Complete analytical validation. Qualify IUO version for Phase 3 use. Establish specimen collection, handling, and processing SOPs	FDA submission of Investigational Device Exemption (IDE) if required for Phase 3 CDx use. Confirm assay lock with FDA before Phase 3 enrolment opens
PHASE 3 — CLINICAL VALIDATION AND PARALLEL SUBMISSION
Phase 3 design	Phase 3 must prospectively use the locked CDx assay (or an analytically bridged version) to define the study population	Final CDx version locked for Phase 3; manufacturing process defined; quality system (21 CFR Part 820 / ISO 13485) in place	Align with FDA on parallel submission plan: NDA/BLA and PMA (or 510(k)) filed simultaneously or NDA/BLA slightly ahead. FDA coordinates review between CDER/CBER and CDRH
Phase 3 conduct	Ensure Phase 3 sites use the IUO/locked CDx assay for enrolment decisions. Collect concordance samples vs local testing where applicable	Support site training, QC monitoring, and specimen logistics for Phase 3. Begin PMA/510(k) preparation: clinical validity section populated with Phase 3 enrolment data	FDA cross-labelling language drafted jointly: drug label references CDx by name and device; CDx label references drug. Begin coordinating with FDA reviewers from both CDER and CDRH
SUBMISSION & APPROVAL — PARALLEL REVIEW
Submission	File NDA/BLA with Module 1 cross-reference to CDx PMA/510(k). Include summary of CDx clinical validation in Module 2.7 (clinical summary) and Module 5 (clinical study reports)	File PMA (for novel CDx) or 510(k) (if predicate exists) simultaneously or within agreed window of NDA/BLA. PMA includes analytical and clinical validation data	CDER and CDRH review simultaneously. FDA coordinates labelling language across both applications. Advisory committee input may be sought for novel biomarker-indication pairs
Approval	Drug approval is contingent on CDx approval. FDA will not approve drug without co-approved CDx if the indication requires biomarker selection	CDx PMA/510(k) approved simultaneously with or just before drug approval. Cross-labelling finalised. Post-market surveillance plan filed	Drug label names the specific approved CDx. Generic alternative assays cannot be used for patient selection without separate FDA approval as an approved CDx for that drug

Approved CDx Examples: The Current Landscape

The table below covers selected approved drug-CDx pairs as of Q4 2025. It illustrates the range of biomarkers, assay types, and CDx partners across oncology. The FDA maintains a complete list of approved CDx devices at fda.gov/medical-devices/in-vitro-diagnostics/list-cleared-or-approved-companion-diagnostic-devices-in-vitro-and-imaging-tools.

Drug (INN)	Company	Indication	CDx Test	CDx Maker	Biomarker
ONCOLOGY — SOLID TUMORS
Osimertinib (Tagrisso)	AstraZeneca	EGFR-mutant NSCLC (1L and 2L)	cobas EGFR Mutation Test v2	Roche	EGFR ex19del / L858R
Sotorasib (Lumakras)	Amgen	KRAS G12C-mutant NSCLC (2L+)	therascreen KRAS RGQ PCR Kit	Qiagen	KRAS G12C
Adagrasib (Krazati)	Mirati/BMS	KRAS G12C-mutant NSCLC (2L+)	therascreen KRAS RGQ PCR Kit	Qiagen	KRAS G12C
Trastuzumab deruxtecan (Enhertu)	AZ / Daiichi	HER2-positive breast, gastric, NSCLC	PATHWAY HER2 IHC / FISH (Ventana)	Roche/Ventana	HER2 IHC/FISH
Pembrolizumab (Keytruda)	Merck	NSCLC 1L (PD-L1 >=50%), multiple solid tumors	PD-L1 IHC 22C3 pharmDx	Agilent/Dako	PD-L1 TPS
Olaparib (Lynparza)	AZ / MSD	BRCA-mutated ovarian, breast, prostate, pancreatic	BRACAnalysis CDx	Myriad Genetics	BRCA1/2 germline
Entrectinib (Rozlytrek)	Roche	NTRK fusion-positive solid tumors	FoundationOne CDx	Foundation Medicine	NTRK1/2/3 fusion
Dabrafenib + Trametinib	Novartis	BRAF V600E/K solid tumors	THxID BRAF Kit	bioMerieux	BRAF V600E/K
HAEMATOLOGIC MALIGNANCIES
Ivosidenib (Tibsovo)	Servier	IDH1-mutant AML	Abbott RealTime IDH1 Assay	Abbott	IDH1 R132
Enasidenib (Idhifa)	BMS	IDH2-mutant AML	Abbott RealTime IDH2 Assay	Abbott	IDH2 R140/R172
Gilteritinib (Xospata)	Astellas	FLT3-mutant AML	LeukoStrat CDx FLT3 Mutation Assay	Invivoscribe	FLT3 ITD/TKD

Selecting a CDx Partner

Build vs Buy vs Partner

Drug sponsors have three options for CDx development: build the assay internally (rare — requires significant IVD regulatory expertise and manufacturing infrastructure); partner with an established IVD company (most common); or work with a reference laboratory that develops a laboratory-developed test (LDT) and pursues FDA approval. The partner model is dominant in oncology, with Roche/Ventana, Agilent/Dako, Foundation Medicine, Qiagen, and Abbott among the most active CDx partners.

Key Partner Selection Criteria

Assay platform fit: the CDx must be technically feasible on a platform the partner has validated and can manufacture at scale. Regulatory track record: CDRH experience matters — partners with prior PMA approvals for CDx in the same biomarker class move faster. Commercial footprint: the CDx must be available in enough CLIA-certified labs to support commercial launch. For global programs, equivalent regulatory experience with CE-IVD (EU), PMDA (Japan), and TGA (Australia) is also relevant.

Co-Development Agreement Structure

The co-development agreement (CDA) between drug sponsor and CDx partner is a complex commercial and legal document. Key terms to negotiate: IP ownership of the biomarker and assay (typically the assay IP stays with the CDx partner; the biomarker IP may be jointly held or licensed); exclusivity (does the CDx partner have exclusivity for the first approved CDx, and for how long); cost sharing (who funds analytical validation, clinical validation, regulatory submissions, and CDRH review fees); and labelling control (who has approval rights over cross-labelling language, which impacts both drug and device labels).

Tip:  Negotiate IP and exclusivity terms before Phase 2 data exists. Once Phase 2 data demonstrates a strong predictive biomarker effect, the CDx partner’s leverage in negotiations increases substantially. Early partnership agreements reached before the data is clear tend to be more balanced.

Tip:  Include a parallel submission commitment in the co-development agreement. Both parties must commit to filing with FDA within an agreed window of each other. A drug NDA filed six months before the PMA is ready creates a review coordination problem that FDA cannot easily solve.

Watch out:  Do not allow local laboratory assays to be used for Phase 3 enrolment without concordance validation against the CDx partner’s locked assay. Local lab variability in biomarker testing is a significant source of Phase 3 enrichment failure and can invalidate the clinical validity argument for the CDx.

Analytical and Clinical Validation: What FDA Expects

Analytical Validation

Demonstrates that the assay accurately and reliably measures the biomarker it is intended to measure. Required studies: accuracy (concordance with a reference method or orthogonal platform), precision (intra- and inter-laboratory reproducibility), analytical sensitivity (limit of detection, particularly for low-frequency alleles in tumour samples), specificity (no false positives from related sequences or mutations), and specimen handling stability (how long samples can be stored before testing without affecting results).

Clinical Validation

Demonstrates that the biomarker-positive/negative classification by the CDx accurately identifies patients who benefit (or do not benefit) from the drug. Clinical validation uses the pivotal Phase 3 trial data — the CDx-defined population must be the population in which efficacy was demonstrated. Key metrics: positive predictive value (PPV), negative predictive value (NPV), sensitivity, and specificity relative to clinical outcome. For oncology CDx, the Phase 3 data itself is typically the clinical validation dataset.

Specimen Type Considerations

The CDx must be validated for the specific specimen type used in clinical practice: FFPE tissue, fresh tissue, liquid biopsy (ctDNA), blood, bone marrow, or CSF depending on the indication. Expanding a CDx from one specimen type to another (e.g. from tissue to liquid biopsy) requires separate analytical and clinical validation studies and a supplemental PMA or new 510(k). Plan specimen type strategy carefully — limiting to tissue only at launch can create access barriers if liquid biopsy becomes standard of care post-approval.

Global CDx Strategy: Beyond the US

EU: CE-IVD Mark under IVDR

In the EU, CDx devices are regulated as Class C IVDs under the In Vitro Diagnostic Regulation (IVDR, Regulation (EU) 2017/746), which replaced the legacy IVDD in May 2022. IVDR introduced a conformity assessment requirement for Class C IVDs through a Notified Body — similar in principle to FDA’s PMA requirement. Notified Body capacity is constrained, and timelines for IVDR certification are running 18-36 months for complex CDx products. Begin Notified Body engagement in parallel with FDA CDRH engagement, not sequentially.

Japan: PMDA CDx Pathway

Japan’s PMDA regulates CDx as medical devices under the Pharmaceutical and Medical Device Act (PMD Act). The PMDA has a CDx co-review pathway similar to FDA’s, with PMDA reviewing the diagnostic device in coordination with the pharmaceutical review division. Japan requires Japanese-specific clinical data for some CDx approvals — confirm whether bridging from US/EU data is acceptable early in development.

Australia: TGA CDx Requirements

The TGA regulates CDx under the IVD regulatory framework. Class 4 IVDs (highest risk, equivalent to Class C under IVDR) require TGA pre-market assessment. Australia participates in the MDSAP (Medical Device Single Audit Program), which allows audit recognition across participating countries. For sponsors with significant Australian clinical trial activity — which the R&D Tax Incentive makes attractive — early TGA engagement on CDx pathway is advisable.

Common CDx Program Failures and How to Avoid Them

Watch out:  Starting CDx co-development after Phase 2 data. The single most common failure mode. By the time Phase 2 data confirms a strong biomarker effect, the CDx assay development timeline cannot be compressed to meet Phase 3 enrolment. Start the CDx partner conversation at IND filing, not at end-of-Phase-2.

Watch out:  Using an unvalidated assay for Phase 3 enrolment. FDA requires that the Phase 3 trial use a version of the CDx assay that has been analytically validated and agreed with FDA before enrolment opens. Retrospective biomarker analysis of Phase 3 samples using the CDx is not a substitute for prospective use — it is treated as exploratory, not as clinical validation.

Watch out:  Failing to plan for site testing logistics. In multi-country Phase 3 trials, ensuring that a consistent CDx assay is available at all sites — or that a central testing laboratory is in place — is a significant operational challenge. Local variability in biomarker testing is the enemy of a clean CDx clinical validation dataset.

Tip:  File a pre-submission (Q-sub) with CDRH before the PMA to align on the analytical and clinical validation plan. CDRH’s Q-sub process is free and non-binding, but the feedback provided is highly predictive of PMA review outcomes. A Q-sub before Phase 3 design is finalised can prevent significant late-stage surprises.

Tip:  Consider FoundationOne CDx or similar pan-tumour NGS panels as a bridging strategy. If your target biomarker is already covered by an approved pan-tumour CDx, FDA may accept labelling that references the panel as the approved CDx, avoiding the need for a bespoke single-analyte assay PMA. Discuss this approach with CDRH and CDER early.

This guide is an editorial product of BioPharmaWire based on publicly available FDA guidance, IVDR regulatory texts, and company disclosures as of Q4 2025. CDx regulatory requirements vary by jurisdiction and are subject to change. This document does not constitute regulatory or legal advice. Drug sponsors and CDx partners should consult qualified regulatory affairs professionals before initiating CDx co-development programs.