Personalized Medicine in Action: The Role of Biomarkers and Companion Diagnostics

Personalized medicine promises to replace one-size-fits-all treatments with therapies matched to each patient's biology. But making that promise real requires two tightly linked tools: biomarkers and companion diagnostics. Biomarkers are measurable indicators of biological states—like a specific gene mutation or protein level—that can predict how a patient will respond to a drug. Companion diagnostics are the tests that detect those biomarkers, often developed alongside the therapy itself. Together, they form the backbone of modern precision oncology and are expanding into other therapeutic areas.

This guide is for anyone who needs to understand how these pieces fit together in practice: researchers designing biomarker studies, clinicians interpreting test results, and biotech professionals evaluating diagnostic partnerships. We will compare different types of biomarkers, explain the development workflow for companion diagnostics, and walk through a realistic example. Along the way, we highlight common mistakes and limitations so you can apply these concepts more effectively.

Why Biomarkers and Companion Diagnostics Matter Now

The shift from empirical medicine to molecularly guided treatment has accelerated over the past decade. Drug developers now routinely identify patient subgroups that benefit most from a new therapy, and regulators expect evidence that the right patients are selected in clinical trials. Without a reliable biomarker and a validated companion diagnostic, a promising drug may fail in a broad population even though it works well in a subset.

For patients, the stakes are direct: a targeted therapy can mean fewer side effects and better outcomes, but only if the diagnostic test accurately identifies who is likely to respond. For healthcare systems, companion diagnostics help avoid wasting resources on ineffective treatments. And for biotech companies, a well-designed diagnostic strategy can accelerate development timelines and improve reimbursement odds.

The Changing Regulatory Landscape

Regulatory agencies like the FDA and EMA have issued guidance on co-development of drugs and companion diagnostics. The trend is toward requiring that the diagnostic be analytically and clinically validated before the drug is approved. This means developers must plan biomarker and test development early, often during preclinical phases. Teams that wait until Phase III to think about the diagnostic often face delays or need to repeat studies.

Economic Pressures and Value-Based Care

Payers increasingly demand evidence that a treatment is cost-effective compared to standard care. A companion diagnostic can demonstrate that the drug works in a defined population, supporting value-based pricing. However, the diagnostic itself adds cost, so the overall benefit must be clear. Many industry surveys suggest that tests with high specificity and sensitivity are more likely to gain coverage, especially when they replace multiple sequential tests.

Core Concepts: Biomarkers and Companion Diagnostics in Plain Language

A biomarker is any measurable characteristic that reflects a biological process, disease state, or response to therapy. In personalized medicine, the most useful biomarkers are predictive—they tell you whether a specific treatment is likely to work. For example, the presence of an EGFR mutation in non-small cell lung cancer predicts response to EGFR tyrosine kinase inhibitors. Companion diagnostics are the assays that measure these biomarkers. They can be based on DNA sequencing, immunohistochemistry, PCR, or other technologies.

Types of Biomarkers

Biomarkers fall into several categories, each with a different role. Prognostic biomarkers indicate the likely course of a disease regardless of treatment. Predictive biomarkers forecast response to a specific therapy. Pharmacodynamic biomarkers show whether a drug is having its intended biological effect. And monitoring biomarkers track disease progression or recurrence over time. In drug development, the predictive type is most often paired with a companion diagnostic.

Companion Diagnostic vs. Complementary Diagnostic

It is important to distinguish companion diagnostics from complementary diagnostics. A companion diagnostic is required for safe and effective use of a drug—without the test, you cannot determine eligibility. A complementary diagnostic identifies a broader population that may benefit, but the drug can still be used without it. For instance, a test for HER2 overexpression is a companion diagnostic for trastuzumab in breast cancer, because only HER2-positive patients are candidates. In contrast, a test for an immunotherapy biomarker like PD-L1 may be complementary if the drug is approved for all patients but works better in those with high expression.

How It Works Under the Hood: Development Workflow

Developing a companion diagnostic alongside a drug involves a structured process that begins with biomarker discovery and ends with regulatory approval and clinical adoption. The workflow typically includes the following stages.

Biomarker Discovery and Analytical Validation

The first step is to identify a candidate biomarker through preclinical research or retrospective analysis of clinical samples. Researchers look for a molecular feature that distinguishes responders from non-responders. Once a candidate is found, an assay is developed to measure it reliably. Analytical validation confirms that the test is accurate, precise, and reproducible across different labs and over time. This includes determining the limit of detection, specificity, and robustness against sample variability.

Clinical Validation and Regulatory Approval

Clinical validation demonstrates that the biomarker actually predicts clinical outcomes. This is typically done by testing the assay on samples from a prospective or retrospective clinical trial. The drug and diagnostic are often co-developed, meaning the trial design includes stratification or enrichment based on the biomarker. Regulators review both the drug and diagnostic data together. For FDA approval, the diagnostic must be cleared or approved as part of the drug label.

Post-Market Surveillance and Test Refinement

After approval, the diagnostic continues to be monitored for performance in real-world populations. Sometimes the test is refined as more data accumulates, or new biomarkers are added to expand the eligible patient group. Laboratories that perform the test must maintain proficiency through external quality assessment programs.

Worked Example: Non-Small Cell Lung Cancer and EGFR Testing

To see how these concepts come together, consider a typical scenario in non-small cell lung cancer (NSCLC). A patient is diagnosed with advanced NSCLC, and the oncologist orders a tumor biopsy for biomarker testing. The standard panel includes EGFR, ALK, ROS1, BRAF, and other actionable mutations. In this example, the test returns positive for an EGFR exon 19 deletion, a predictive biomarker for response to EGFR tyrosine kinase inhibitors like osimertinib.

Decision Points and Trade-offs

The oncologist now has a clear treatment path: prescribe osimertinib. But several decisions arise. Should the test be done on tissue biopsy or liquid biopsy? Tissue is the gold standard, but liquid biopsy (ctDNA from blood) can detect mutations when tissue is insufficient. However, liquid biopsy has lower sensitivity for some alterations. In this case, the tissue test was positive, so no further testing is needed. Another trade-off is turnaround time: some labs return results in 3-5 days, while others take two weeks. For a patient with rapid disease progression, a faster test may be worth the extra cost.

What Could Go Wrong

Common pitfalls include false negatives due to poor sample quality or low tumor content. If the biopsy contains mostly normal cells, a mutation may be missed. Some teams use reflex testing: if the initial test is negative, they automatically proceed to a broader panel or liquid biopsy. Another issue is variant interpretation: not all EGFR mutations are equally sensitive to osimertinib. The exon 19 deletion is highly sensitive, but some rare mutations may confer resistance. Laboratories must report variants with clear clinical significance annotations.

Edge Cases and Exceptions

Not all biomarkers are straightforward. Some are continuous rather than binary, requiring a cutoff value that can be debated. For example, PD-L1 expression is measured as a percentage of positive tumor cells, and different drugs use different cutoffs (e.g., 50% for pembrolizumab first-line, 1% for second-line). This creates confusion for clinicians and labs. Another edge case is tumor heterogeneity: a biopsy from one site may not represent the entire tumor, leading to discordant results. Repeat biopsies or multi-region sampling can help but are not always feasible.

Emerging Biomarkers: Tumor Mutational Burden and Microsatellite Instability

Some biomarkers are agnostic to tumor type. Tumor mutational burden (TMB) and microsatellite instability (MSI) are used to identify patients likely to respond to immune checkpoint inhibitors across multiple cancer types. However, TMB measurement varies by panel and lacks a universal cutoff. MSI testing is more standardized but still requires careful validation. These pan-tumor biomarkers expand the reach of personalized medicine but add complexity to laboratory workflows.

Rare Mutations and Off-Label Use

When a patient has a rare mutation for which a drug is approved in a different cancer type, clinicians may consider off-label use. Some biomarkers, like NTRK fusions, are actionable across histologies. But payers may not cover the drug without a companion diagnostic specific to that indication. In such cases, clinical trials or compassionate use programs are often the only route.

Limits of the Approach

Despite its power, biomarker-guided personalized medicine has real limitations. First, not all diseases have known predictive biomarkers. For many conditions, the molecular drivers are not yet understood, or the biomarkers are not clinically actionable. Second, the cost of developing companion diagnostics can be high, and the tests may not be available in all regions, creating disparities in access. Third, tumor evolution can lead to acquired resistance: a patient who initially responds to a targeted therapy may develop new mutations that render the drug ineffective. Repeat biopsies and monitoring are needed, but they are not always performed.

Technical and Logistical Hurdles

Laboratory standardization remains a challenge. Different assays for the same biomarker can produce discordant results, especially for complex biomarkers like TMB or PD-L1. External quality assurance programs help, but variability persists. Additionally, turnaround times can delay treatment decisions. Some hospitals have implemented rapid molecular testing programs to address this, but resource constraints limit adoption.

Regulatory and Reimbursement Barriers

Even when a companion diagnostic is approved, reimbursement may lag. Payers may require additional evidence of clinical utility or cost-effectiveness. For multi-gene panels, individual tests may not be covered, forcing labs to bill each component separately. This administrative burden can discourage adoption. Advocacy for streamlined coverage policies is ongoing, but progress is slow.

As personalized medicine continues to evolve, the role of biomarkers and companion diagnostics will only grow. For now, the most practical advice is to invest in robust analytical validation, plan the diagnostic strategy early in drug development, and stay informed about evolving regulatory expectations. Teams that treat the diagnostic as an afterthought often find themselves redoing work or missing market opportunities. By understanding both the power and the limitations of these tools, you can make better decisions for patients and for your organization.