Precision medicine has changed the way doctors prescribe treatments. Instead of using a one-size-fits-all approach, clinicians increasingly tailor therapies to individual patients. Companion diagnostics sit at the center of this shift. These tests help doctors determine whether a patient is likely to benefit from a specific drug, face a higher risk of adverse reactions, or require adjustments to therapy.
To support the safe use of these tests, Taiwan’s regulator has issued technical guidance for in vitro companion diagnostic devices (IVD-CDx). The guidance outlines what manufacturers must show when they apply for device registration. It also clarifies how companies should evaluate analytical performance and clinical effectiveness before bringing a companion diagnostic to market.
What Is a Companion Diagnostic?
A companion diagnostic is an in vitro test that provides information essential for the safe and effective use of a particular drug. Unlike standard diagnostic tests that simply detect disease, these tests connect directly to a specific therapy. In other words, they help clinicians decide who should receive a drug and how that drug should be used. Most companion diagnostics serve one of three purposes.
First, they identify patients who are most likely to benefit from a treatment. Many targeted therapies only work in patients who carry certain genetic mutations or biomarkers. Without a diagnostic test, prescribing these drugs would resemble throwing darts in the dark.
Second, companion diagnostics can flag patients who face a higher risk of serious side effects. Some biomarkers predict toxic responses to specific medicines. Detecting those markers allows doctors to avoid treatments that might do more harm than good.
Third, clinicians sometimes use these tests to monitor treatment response and optimize therapy. For example, test results might guide dosing decisions, signal when a treatment stops working, or indicate when a physician should discontinue therapy.
Taken together, companion diagnostics help clinicians match the right treatment to the right patient at the right time.
Regulatory Framework
Manufacturers developing companion diagnostics must comply with Taiwan’s medical device registration rules as well as requirements for in vitro diagnostic devices. The new guidance adds another layer of detail. It explains what regulators expect to see in performance studies and clinical evidence.
Regulators drew heavily from international standards, including guidance from the United States, the European Union, and Japan, as well as recommendations from the Asian Harmonization Working Party (AHWP).
Required Information for Registration
A strong submission begins with clear documentation. Manufacturers must describe the device’s structure, materials, specifications, and performance characteristics. However, the most critical element is the intended use statement.
This statement should clearly identify the biomarker or analyte detected by the test. For many companion diagnostics, the target involves a specific gene mutation or protein marker. If the device detects a particular genotype, the manufacturer should also describe the gene sequence and major mutation regions involved.
The intended use must also explain how the test connects to the associated drug. Does the test select eligible patients? Does it identify safety risks? Or does it monitor treatment response? Regulators expect manufacturers to name the corresponding drug or active ingredient as well.
Manufacturers must also clarify whether the assay produces qualitative, semi-quantitative, or quantitative results. That distinction matters because it affects how clinicians interpret test outcomes.
Beyond the intended use, submissions should describe the full testing process. This includes the types of specimens used, such as serum, plasma, tissue samples, or urine, and the patient population for which the test is designed.
Equally important, manufacturers must explain the testing methodology. They should outline how the assay works, what instruments it requires, and whether the system runs automatically.
A submission should also list all components involved in the assay. These may include antibodies, antigens, nucleic acid primers, buffers, substrates, calibrators, and quality control materials. In addition, manufacturers must explain how laboratories should collect, handle, transport, and store specimens.
Finally, if the device relies on specialized software or includes accessories or multiple packaging configurations, the submission should describe those elements as well.
Analytical and Preclinical Testing
Before companies evaluate a companion diagnostic in patients, they must first confirm that it works reliably in controlled laboratory settings. These early studies, often called analytical or preclinical testing, serve as the technical foundation for the device’s regulatory submission.
At this stage, developers assess whether the test can accurately detect the intended biomarker and avoid producing false signals. They also examine whether outside factors in biological samples could interfere with results. In addition, manufacturers confirm that the test produces consistent results when repeated or when performed in different laboratories.
Researchers must also define the threshold that separates positive from negative results and ensure that the test remains stable during storage and routine use. If the device can analyze different types of specimens, such as blood or tissue, each type must be evaluated separately. Together, these studies demonstrate that the test performs reliably before it moves into clinical research.
Comparative Testing
Whenever possible, manufacturers should compare their device with similar products already approved in major markets such as the United States, Japan, Canada, Switzerland, Australia, or the European Union.
However, this approach does not always work. Novel technologies sometimes lack a direct competitor. In those cases, companies may compare the assay with a recognized reference method or rely on internationally accepted testing standards.
Either way, the goal remains the same: show that the new device performs as well as, or better than, existing methods.
Clinical Performance Studies
Analytical performance alone does not guarantee clinical usefulness. Manufacturers must also show that the test works in real-world clinical settings.
To do this, developers design studies that evaluate how accurately the diagnostic identifies the intended patient population. Sample size calculations should reflect disease prevalence and the intended use of the device.
Randomized controlled trials often provide the strongest evidence. However, regulators may accept other study designs when appropriate. For example, enrichment designs allow researchers to select patients based on biomarker status. Another option involves analyzing leftover specimens from drug clinical trials in a prospective-retrospective study.
In certain situations, single-arm studies may also suffice. This approach can work when patient numbers remain small or when drug response rates serve as the primary endpoint.
Clinical performance typically relies on measures such as positive percent agreement (PPA) and negative percent agreement (NPA). These statistics show how closely the new diagnostic aligns with a reference method or comparator.
Coordinating with Drug Development
Companion diagnostics rarely stand alone. Their success depends on the therapy they support. Because of this connection, manufacturers should coordinate diagnostic development with the drug development timeline.
When both products move through development together, researchers can evaluate the diagnostic using specimens from the same clinical trials that test the drug. This approach strengthens the evidence linking the biomarker to treatment response.
Final Thoughts
By allowing doctors to make better-informed decisions, companion diagnostic tests play an increasingly important role in precision medicine. Taiwan’s guidelines provide a clear path for manufacturers to bring their products to market by identifying what type of evidence must be provided to demonstrate the reliability of a diagnostic test, as well as what biomarker the diagnostic test is intended to detect, and how it relates to the therapeutic use of the device. Furthermore, the guidelines adopted by Taiwan mirror many of the current international regulatory approaches. Understanding these expectations early in the process can improve the efficiency of companion diagnostic test registration and allow for more timely access to these tests for patients.
Written by: Ames Gross – President and Founder, Pacific Bridge Medical (PBM)
Mr. Gross founded PBM in 1988 and has helped hundreds of medical companies with regulatory and business development issues in Asia. He is recognized nationally and internationally as a leader in the Asian medical markets. Mr. Gross has a BA degree, Phi Beta Kappa, from the University of Pennsylvania and an MBA from Columbia University.