For much of the history of oncology, cancer treatment was primarily based on the location of the tumor’s origin. A patient with breast cancer, lung cancer, colorectal cancer, or melanoma would receive treatment tailored to the specific type of cancer. While this approach provided valuable insights and remains relevant, it may not be enough in today’s complex medical landscape.
Two patients diagnosed with the same cancer at the same stage can respond very differently to the same treatment. One patient may experience a dramatic and long-lasting response, while the other may receive little to no benefit. Part of the answer lies in the underlying biology that contributes to these differences.
Over the years I’ve spent in oncology clinical research and drug development, this has been one of the most important shifts I’ve watched happen. We’ve moved from asking primarily, “Where did this cancer start?” to also asking: “What is driving it?” This sounds like a subtle change, but it isn’t. It has transformed how we classify some cancers, how we design clinical trials, how we develop drugs, and ultimately how we decide which treatment may be most appropriate for an individual patient.Biomarkers are at the heart of this transformation.
A Cancer Diagnosis Is Only the Beginning of the Story
When most people hear the word “cancer,” they naturally think about the organ involved. For instance, lung cancer starts in the lung, breast cancer starts in the breast, and colon cancer starts in the colon. But underneath these anatomical labels is an enormous amount of biological variation.
Cancer develops because normal cellular processes have gone wrong. Genetic alterations, abnormal proteins, disrupted signaling pathways, immune interactions, and many other biological changes can contribute to the growth and survival of a tumor. Importantly, these changes vary significantly between patients’ cancers.
That’s why two tumors originating in the same organ may behave very differently. Biomarker testing helps us look beneath the diagnosis. It gives us information about the characteristics of the tumor, and sometimes the patient, that may help predict how the cancer will behave or whether a particular treatment is more likely to work. In some cases, this information can completely change the treatment strategy.
Biomarkers Have Changed What “The Right Treatment” Means
Earlier generations of cancer therapy heavily relied on treatments that targeted rapidly dividing cells. While these therapies have saved countless lives and remain incredibly important, they were not specifically designed around a specific biological vulnerability within an individual tumor.
However, targeted therapy revolutionized this perspective. If a tumor relies on a specific molecular pathway or carriers a particular alteration, could we develop a therapy that disrupts it? In certain cancers, the answer has been yes. We now have therapies that are specifically designed to target alterations or specific targets such as HER2, EGFR, ALK, BRAF, RET, NTRK, KRAS, and numerous others.
Some biomarkers help us identify patients who may benefit from targeted therapies, while others help guide immunotherapy. Additionally, they provide information about prognosis, resistance, or how a cancer may evolve. And we’re increasingly using combinations of biomarkers rather than relying solely on one result. That means the question is no longer simply asking, “Which drugs are approved for this type of cancer?” It’s asking, “What do we know about this particular cancer that can help us choose among them?” This approach is known as precision oncology.
Sometimes the Same Biomarker Matters Across Different Cancers
One of the most fascinating developments in oncology has been the realization that the biology of a tumor can sometimes matter as much as, or even more than, the organ where it originated. A molecular alteration may occur in several different tumor types. This has helped create an entirely different way of developing some cancer therapies.
Instead of studying a treatment only in lung cancer or only in colorectal cancer, researchers may now study patients whose tumors share a particular molecular characteristic. This is one reason we’ve seen growing interest in tumor-agnostic treatment strategies and basket trials. It represents an important shift in how we think about cancer, where the organ is no longer the sole determinant of treatment, but biology also plays a crucial role.
But Finding a Biomarker Doesn’t Automatically Mean We Have the Answer
This is where precision oncology becomes more complicated than it sometimes sounds. Finding an alteration in a tumor doesn’t necessarily mean that alteration is driving the cancer. And even when it’s important biologically, we may not have a therapy capable of targeting it effectively. Sometimes we have a drug, but the patient has already developed a resistance mechanism. Sometimes the same biomarker behaves differently depending on the tumor type or the other molecular changes surrounding it.
Cancer biology rarely involves a single mutation, a single drug, or a single response. I think that’s important to acknowledge because the language around personalized medicine can sometimes make it sound much simpler than it really is. Biomarker testing provides us with additional information, but the true value lies in our ability to interpret that information within the clinical context.
The Test Itself Matters
This is something we see constantly in clinical development. People talk about “biomarker testing” as though it were one universal test and it’s not. Different biomarkers may require different methods. Some can be evaluated through tumor tissue, some through blood, and some require DNA or RNA sequencing. Others are based on protein expression or other characteristics.
Even when two tests are looking for the same general biomarker, the methodology, sensitivity, specimen requirements, and thresholds may vary. Additionally, the age and quality of the tissue, as well as whether the sample was collected before or after a specific therapy, can also play a role.
This is especially important in clinical trials. If a study requires a particular biomarker, we have to know exactly what qualifies. Can local testing be accepted? Is central confirmation required? Is archival tissue adequate? Does the patient need another biopsy? What happens if the local and central results disagree? These may sound like technical details, but for the patient waiting for treatment, they hold immense significance.
Precision Medicine Has an Operational Side We Don’t Talk About Enough
This is one of the lessons drug development has taught me. You can have a brilliant biomarker strategy on paper and still struggle to execute it. Imagine a patient with advanced cancer who may qualify for a biomarker-driven clinical trial. First, the right test has to be ordered. Then an adequate sample has to be available. If it isn’t, the patient may need another biopsy. The sample has to reach the laboratory, testing has to be completed, the result has to come back, someone has to interpret it, eligibility has to be confirmed, and all of this has to happen while the patient’s cancer is still progressing. This timeline is so important.
A biomarker result that arrives too late may be scientifically accurate but clinically useless for that particular treatment decision. This is why I’ve come to view biomarker strategy and operational strategy as inseparable. If precision medicine depends on getting the right information at the right time, then turnaround time isn’t simply a laboratory metric; it becomes a clinical issue.
Patient Selection Can Determine What We Learn About a Drug
Biomarkers have also fundamentally changed clinical trial design. If we believe a therapy works primarily in tumors with a particular characteristic, then enrolling the right population becomes critical. Enrolling too broadly can dilute a meaningful efficacy signal, while selecting too narrowly may exclude patients who could benefit. Choosing the wrong biomarker threshold can significantly impact the entire development strategy.
This is why patient selection is one of the most crucial decisions in oncology drug development not just because it involves finding patients for trials, but also because it ensures that the trials address the right biological questions. A drug can be effective and still look disappointing if we study it in the wrong population. This is a lesson our industry has learned repeatedly.
Biomarkers Can Also Tell Us Why Treatment Stops Working
Precision oncology isn’t only about selecting the first treatment. Cancer changes and under the influence of therapy, resistant populations can emerge. A tumor that initially depends on one pathway may develop another way to survive. This means the biomarker profile we see at diagnosis may not tell us the entire story several years later. This is why repeat molecular testing, circulating tumor DNA, and other approaches to monitoring tumor evolution are gaining significant interest.
The future of cancer treatment may be much more dynamic. Instead of making a single decision at diagnosis, precision medicine involves an ongoing understanding of the biology of the disease. This approach involves testing, treating, monitoring, and understanding how the cancer changes, allowing for changes as needed.
Biomarkers Don’t Replace Clinical Judgment
This is something I feel strongly about. A genomic report can tell us a lot, but we still have to decide what matters for the patient in front of us. Not every finding is actionable. Even when it is, a treatment that matches an alteration may not be the right choice for that patient at that point in their care. We have to consider their disease, prior treatments, comorbidities, performance status, available options, potential toxicity, and what they hope to achieve with treatment. A biomarker can be a crucial part of that decision, but it’s still part of a bigger clinical picture.
The Future Will Go Beyond One Biomarker and One Drug
I think this is where oncology becomes particularly exciting. We’re moving toward a much deeper understanding of cancer. This involves genomics, proteomics, immune profiling, circulating tumor DNA, minimal residual disease, resistance mechanisms, and potentially multiple layers of biological information analyzed together.
Artificial intelligence may eventually help us interpret increasingly complex combinations of clinical and molecular data. And treatments themselves are becoming more sophisticated, including targeted therapies, antibody-drug conjugates, bispecific antibodies, cell therapies, personalized cancer vaccines, and gene-engineered approaches.
The future of precision medicine may be less about matching a single biomarker to a drug and more about understanding the biology of a patient’s cancer as a whole, and using that understanding to choose the best treatment at that point in their care. This is a much more ambitious goal.
Access May Become One of the Biggest Challenges
There is one more part of this conversation we shouldn’t ignore. Precision medicine only works if patients can access it. Access to comprehensive biomarker testing isn’t uniform. Neither is access to academic cancer centers, molecular tumor boards, specialized laboratories, or clinical trials.
If we’re developing precise therapies but only a fraction of eligible patients are ever tested for the biomarkers that identify them, then we’ve solved only part of the problem. The next phase of precision oncology can’t simply be about discovering more biomarkers. It has to be about making meaningful testing available to the patients who may benefit from it.
A Final Thought
When I think about where oncology was when I began my career and where we are today, biomarker-driven treatment is one of the changes that stands out most. We’ve moved beyond asking only what a cancer looks like and where it originated. We’re now asking what makes it behave the way it does, what drives its growth, what makes it vulnerable, why it responds, why it becomes resistant, and what that information can teach us about the treatment we choose next.
We still have a long way to go. We won’t find an actionable biomarker for every patient, and even when we do, the matched treatment may not work as we hope. Resistance remains a major challenge, access to testing is uneven, and cancer biology will continue to surprise us. Still, I think the direction is clear. As oncology becomes more precise, progress will mean more than having more targeted drugs. It will mean knowing who is likely to benefit from them, when to use them, and why.
To me, this is the real promise of biomarker testing. It doesn’t just help us understand the cancer better. It helps us move closer to a question we’ve been trying to answer throughout the history of oncology: What is the right treatment for the patient sitting in front of us? The more we understand the biology behind each cancer, the better chance we have of answering that question well.