The Companion Diagnostic Risk: When Your Drug Depends on a Test You Don't Control

← Back to All Insights

Precision medicine has made diagnostic tests part of many drug approvals. In 2024, FDA approved 18 personalized medicines, about 38 percent of newly approved therapeutic molecular entities, by the Personalized Medicine Coalition's count, and the Coalition finds personalized medicines have made up at least a quarter of new drug approvals in each of the last ten years. When a drug's labeling depends on a test, the test becomes part of the asset, and diligence that examines only the drug's clinical and regulatory profile can miss half of the picture.

What Makes a Companion Diagnostic Different

FDA describes a companion diagnostic as a medical device, often an in vitro diagnostic, that provides information essential for the safe and effective use of a corresponding drug or biological product. Under FDA's 2014 guidance, the test and the drug should in most circumstances be approved or cleared at the same time, and use of the test is stipulated in the drug's labeling; for a novel drug that needs one, FDA generally will not approve the drug if the companion diagnostic is not approved or cleared for that indication. The guidance expects most companion diagnostics to be Class III devices reviewed through a PMA, with some cases where a Class II device cleared through a 510(k) is appropriate, and FDA's list of authorized companion diagnostics includes 510(k) and De Novo authorizations alongside PMAs.

Single-Test Dependence

A drug may launch with one authorized companion diagnostic from one manufacturer, sometimes run at a limited number of laboratories. If that test's availability, capacity or commercial support changes, identifying eligible patients can slow. FDA's guidance recommends that drug labeling specify use of an FDA-approved or cleared companion diagnostic of the type described, rather than one manufacturer's test, so that more than one test can be used, and FDA's 2020 guidance on labeling companion diagnostics for a group of oncology drugs supports claims that cover several therapies rather than a single product. Whether a second test exists or is in development is a fair diligence question.

Laboratory-Developed Tests

FDA's 2024 rule that would have phased laboratory-developed tests (LDTs) into device oversight was vacated by a federal court on March 31, 2025, and in September 2025 FDA restored the prior regulation text, so LDTs remain under CLIA. Where testing for a drug's biomarker runs partly through LDTs, patients may be selected with laboratory tests that have not had FDA premarket review alongside the authorized companion diagnostic, and it is worth knowing how those results compare. See our article on the LDT rule.

Test Coverage and Payer Variability

Coverage for the test is a separate decision from coverage for the drug. In Medicare, molecular tests billed in the jurisdictions that participate in the MolDX program follow that program's requirements, while commercial payers set their own coverage policies and prior authorization rules. Uneven coverage for the test can slow uptake of a drug even where the drug itself is covered.

Diligence Questions

Where a therapeutic depends on a companion diagnostic, questions worth asking include: how many authorized tests support the indication, and from how many manufacturers; how the diagnostic partner's commitment to the indication is secured, and for how long; how coverage for the test varies across payers; whether the biomarker can be measured by more than one method, and whether those methods agree; and whether the drug and the test have different owners whose incentives could diverge.

Companion diagnostic dependency is not a deal-breaker in itself. It is a structural feature of precision medicine that deserves its own lane in diligence, alongside the drug's clinical and regulatory profile.