Biomarker & Companion Diagnostics: a 4-Part Primer
#1. The Importance of Diagnostic Tests in Predicting Treatment Benefits
Diagnostic tests are the basis of 85% of the decisions made in clinical medicine. There are well over 3,000 commonly available lab tests in use today, and they fall into one of 4 general categories:
1. Diagnosis (to rule in or rule out a diagnosis)
2. Monitoring (e.g., the effect of drug therapy)
3. Screening (e.g., for evidence of possible disease)
4. Research (for use in clinical trials)
Yet, when it comes to choosing specific targeted therapies for neoplastic disease, physicians are often left in the dark. In this new era of “personalized medicine” the way targeted therapies are matched to the appropriate patients who should receive them is through the use of diagnostic tests for a biomarker. This opens the door for a fifth category:
5. Companion diagnostics
#2. Personalized Medicine & the Role of Predictive vs. Prognostic Markers
There has been some confusion about the definitions of the terms prognostic and predictive biomarkers. The original definition of a prognostic biomarker is one that provides information on the likely course of the cancer disease in an untreated individual, or more recently, the likely course of disease regardless of treatment. Especially in low risk, early disease, these markers can guide physicians by identifying patients who may not need post surgical cytotoxic chemotherapy.
In contrast, predictive markers are those that identify patients who are appropriate for specific targeted therapies. They predict the likelihood of benefit that patients may receive from these drugs. Utility of predictive markers must be established in well designed prospective trials. The term “Precision Medicine” is built around this concept. The best examples of this are listed in the table above. As research into targeted therapies unfolds, the list will grow.
#3. Biomarker Sensitivity and Specificity and Test Result Accuracy
In this new age of targeted therapy, thousands of putative biomarkers have been identified and published. We see the posters at every scientific meeting. These have dramatically increased the opportunities for developing more effective therapeutics. However, the transfer of biomarkers from discovery to clinical practice is still a process filled with lots of pitfalls and limitations. To become a clinically approved test, a potential biomarker should be confirmed and validated using hundreds of specimens and should be reproducible, specific and sensitive.
These two terms are often confusing and sometimes used interchangeably. There are no 100% sensitive and specific biomarkers for any tumor type to date. A biomarker with a high sensitivity has a low specificity and vice versa. Unfortunately, biomarkers with ideal specificity and sensitivity are difficult to find.
Sensitivity is the ability of a test to accurately identify a patient/tumor sample that has the characterstic for which it is being tested, For example, breast cancer patients who test positive for HER2 actually have an abundance of HER2 receptors on their tumor cells. Tests that report HER2 positivity when, in fact the patient does not have an abundance of HER2 receptors are called false positive.
Specificity, on the other hand is the ability of a test to accurately identify those who do not have the characteristic being tested for. In the HER2 example, patients testing negative for HER2 do not have overexpression of the HER2 receptor on their tumor cells. Therefore, tests reporting HER2 negative when, if fact there is HER2 overexpression, are called false negative.
Sensitivity and specificity are characteristics of the testing method, and are expressed in percentage. They confer confidence in the accuracy of the test results.
#4. Biomarker Evaluation and Analysis: Determining True Utility
Now that drug developers have acknowledged the importance of biomarkers in their efforts to move candidates into the market, there is a rush to find ones that can successfully predict benefit. Because test developers do not fully understand the drug development process, and pharma companies do not always fully understand diagnostic testing, some markers are not clinically useful.
Finding a biomarker that predicts the effects of treatment on the patient is a daunting task. There have been many misguided efforts to require biomarker testing as a requirement of drug treatment. One example is the requirement that EGFR testing be performed on tumor samples prior to treatment with EGFR MAbs like Erbitux and Vectibix. It turns out that testing for the EGFR receptors is irrelevant and is neither prognostic or predictive of treatment outcome. A second example is the use of KRAS in NSCLC. It is well known that KRAS mutations are predictive for lack of effectiveness of EGFR MAbs in colorectal cancer. Many oncologists also test NSCLC patients for KRAS, but it cannot be used to guide treatment as it is only a prognostic marker for that indication.
There are many other examples of biomarkers that, while touted as possible predictive markers by drug developers, their true clinical utility remains in question.
Armed with a clearer understanding of issues surrounding biomarker development, Pennside can boost your Competitive Advantage by providing you with the landscape of diagnostic testing that is integral to targeted drug development today.
As part of Pennside’s biomarker analysis service, we can help you formulate a powerful strategy for biomarker development programs for targeted therapy companion diagnostics.